Datisto application handbook

1. Explicitly choose a year and accounts

In Portfolio, open realized sales and the Annual sales report. Enter a year from 0001–9998 in Calendar year · UTC and select one to 100 own selling accounts. All accounts and Clear accounts change only the draft. Select Apply year and accounts. The year follows the sale’s original effectiveAt timestamp in UTC; draft edits do not replace an already applied report.

The report uses the complete admitted financial ledger and its displayed economic cutoff. Accounts form your reporting selection. This is not a legally verified tax group or a merger of acquisition lots across accounts.

2. Follow each original sale and consumed lot

Each sale retains its original quantity, proceeds, acquisition basis, PnL and consumed lots. The existing FIFO within each account is retained; selecting accounts does not run a second FIFO calculation. A matching instrument identifier does not authorize matching sales across accounts.

A sale’s account can differ from the original acquisition account after a transfer. A related acquisition may therefore come from an account outside your selection. Native currencies stay separate; portfolio-base-currency amounts have their own columns and availability. Lots and sources opens the consumed lots and a route to the original acquisition event.

3. Read the quality of individual inputs

Missing acquisition basis, proceeds, FX or acquisition date is shown as unknown with a reason. Zero is a known value, not a replacement for missing data. Totals are separated by currency, and an individual total column remains unknown if any contributing value is missing. One available column does not verify the others.

Also inspect source issues and calendar-period coverage. If the displayed economic cutoff does not cover the end of the selected year, the period is partial. A year that has not started at that cutoff is not a forecast of future sales. Even a full calendar period and known basis do not establish complete broker history. The report does not verify tax residence, rates, holding-period tests, exemptions or a tax return.

4. Renew access to the same source

An applied report retains the complete ledger’s exact revision and fingerprint, year, accounts and economic cutoff. Verify the same report or export rereads that same source at the same cutoff. It does not silently replace it with the latest ledger or a different cutoff.

If your account, session, source or availability changes, verify your own portfolio again. To use a different source or selection, explicitly load and apply it. Reading and exporting do not change financial events.

5. Download the whole evidence packet, not a table page

Whole report JSON and Whole report CSV include the whole applied report, all pages, selection, totals, gaps and consumed-lot provenance. They also include all original own input events and accounts of that portfolio once, including unselected accounts, so the calculation’s source can be inspected. The selection scope and complete input scope are distinguished in the export. Keep the file private.

The export omits write idempotency keys, owner or session identities and authentication secrets. It is a readable evidence packet, not a tax calculation, import or automatically restorable backup. The separate original-ledger export is available in Activity; unknown derived amounts do not mean the original records are missing.

6. Save and manage your own account groups

Under Named account groups, save the current draft of one to 100 own accounts under a name. The library holds up to 50 groups. Select Review group save, inspect the change and whole collection in the request, then Confirm account groups save. Only the name, portfolio and exact account identities are saved; the year, amounts, lots and report results are not.

Verify groups reads your current own collection. Find a group and inspect its accounts and exact identities. Adopt into selection requires the same portfolio and every account in the currently verified own source. A missing or different account is not substituted or silently skipped; no partial selection is adopted. Adoption changes only the account draft. The year and last applied report remain; a new report requires Apply year and accounts.

Rename changes the group’s name. Review replacement with current draft proposes replacing its members with the current draft selection from the same portfolio. Review deletion removes only the library group, not an account or its events. Each change requires review and separate confirmation of the whole collection; Cancel before sending saves nothing. A group is not a tax pool and does not change FIFO within accounts.

If the collection has a newer revision, cancel the unsent draft, verify the current list and prepare the change again. If a command was already sent and its outcome is uncertain, its original form remains retained. Verify and retry the same command uses the original key and content; nothing retries automatically. Reloading the tab can recover the original command when session storage works; follow the displayed notice for memory-only recovery. Confirmation of an older command does not roll back a newer collection. After an account or session change, verify the same own account first; saved groups themselves grant no financial-data access.

Save the original report in My analyses

Apply prepares one actual private result from the complete owned ledger. Save Snapshot opens a review of its name and original result; separate confirmation saves it in My analyses. The whole original report and inputs needed for FIFO are retained, not merely the current table page. Later transactions or portfolio renaming do not recalculate the original snapshot.

Reopen or prepare a new report

Open the original annual report from My analyses. Verification rechecks current access to the same snapshot; JSON and CSV belong to that result. To calculate anew, choose to adopt its selection in Portfolio, review actual current accounts and explicitly confirm any draft replacement. Adoption alone changes no results; another Apply creates the new report. The snapshot remains private and establishes neither a tax calculation nor complete broker history.

Private archive of stored own inputs

Sign in and load your export jobs. Review the scope, give consent and confirm one request. After confirmed queue acceptance, preparation continues in the background even after leaving. Refresh status manually. For an uncertain outcome, check the original command first; explicitly retrying the same key never creates a second job.

Extract the ready TAR.GZ: manifest.json lists every section, column, exclusion and fingerprint. JSONL has one original object per line; CSV has one record per row and fixed section columns. Amounts remain exact strings. CSV prefixes formula-like text with an apostrophe and displays null as blank; JSONL preserves original values including null versus empty text. Data file and byte counts exclude the manifest itself.

All parts come from one coherent read when the job is processed, not necessarily from the button-click time. Passwords, tokens, MFA secrets, foreign messages and provider results are excluded. Your references grant no source rights. Purged original imports and deleted words are not reconstructed. The manifest discloses the separate CMS, external copies and other exclusions. This is not a restorable whole-application backup or complete handling of all personal-data portability obligations.

Only the current signed-in owner may download the archive; technical delivery expiry is 24 hours from the request. Every download rechecks account, session, job expiry and file fingerprints. After expiry, request a new current snapshot. This period neither erases source inputs nor defines general legal account retention. Incomplete preparation is not issued as a complete archive. Actual account erasure has its own review and separate steps described below.

An actual account erasure request

Below the actual export, review an actual erasure request. Download your own data if needed and retain the exact recovery JSON outside this page first. It contains only a random code and command identity; it reads progress after sign-out and cannot sign in or initiate another erasure. Handing the file to the browser does not establish that you saved it.

Confirm you retained the code, type DELETE ACCOUNT exactly and enter your current password. Confirm with the separate button. An account using multifactor verification or a staff role also requires recent verification in Account security. The password is cleared after sending and never retained for retries. Acceptance revokes all sessions, blocks new sign-ins and withdraws own public words and sharing links.

Ordinary own application data is then erased in the background. Other people’s replies are preserved; where they need their original context, empty structure remains without the original author or words. Restricted security and moderation records, the publishing system, external services and backups have their own review-required steps. Application data processed does not establish physical deletion of every copy or a legal retention period.

If a response is lost, do not send a replacement request. Check the original status with the retained code only. After leaving the page, paste recovery JSON in the separate recovery area, load it without sending, then explicitly check progress. This needs no sign-in and never sends erasure. A missing record does not prove that an in-flight write cannot commit. Cancelling the wait never revokes an already accepted request.

Remaining-copy review for administrators

An authorized administrator opens Review remaining account copies in Settings. Access requires an active administrator role, verified email, enabled and enrolled multifactor authentication and recent verification in Account security. Ordinary sign-in or knowing a request identifier grants no access.

After explicit verification, page through the queue and open a request’s exact inventory. It contains original internal references, not copied messages or the private recovery code. The original subject identifier remains personal data. The fingerprint identifies the recorded inventory; it does not establish that a copy in another system still exists or has been removed.

Open a review for one original reference. Inspect the current record and paginated history, then prepare either a provisional hold for further review or a required follow-up action. Enter your own reason and an opaque reference to related evidence; do not paste case contents, passwords or keys. Review the whole draft and save with a separate confirmation. A newer decision by another administrator requires a fresh review.

The receipt preserves the original decision while the current review may change later. If a response is lost, check the original command first. Nothing retries automatically; an explicit retry uses exactly the same key and content. A missing record does not establish cancellation of an in-flight write. Recovery after a new sign-in by the same administrator requires fresh verification; the pending command lives only in this page’s memory.

A review record alone neither erases nor releases a copy and changes no account-erasure completion state. The optional next-review time is a recorded field only: it schedules nothing, ends no hold and establishes no legal deadline. An unrecorded inventory is unavailable, not proof of zero copies. Sensitive details are hidden when the short verification expires or the page is hidden; load them again through explicit verification.

After application erasure, explicitly verify allowed fields for one supported exact audit reference. The preview shows the event type, whole-record fingerprint and field names, never original personal values. A staff-binding change supports the original operator’s reason and identifier; a moderation or correction decision supports only the original reason. Unknown types and inconsistent associations are unavailable for this action. Reading this preview records a minimal access audit without copying original words.

Actual removal of selected fields requires a current Action required review. Choose fields, enter an opaque approval reference and inspect the exact draft. Separate confirmation is bound to the original source fingerprint and review revision; a hold blocks a new action. The receipt preserves removed field names, before/after fingerprints, the review revision and time, never removed text. History and original receipts support recovery after a lost response without repeating the change. Other event fields, case records, CMS, external copies and backups are not resolved by this step.

What you can download today

In Settings, open My data and privacy. Download the six current browser preferences directly: language, theme, density, currency, time zone and sign-in destination. The file contains no sign-in, password, token, profile or portfolio. It is not a complete account backup and is not imported automatically.

The overview leads to exports of a selected own portfolio and the investment journal’s current own words. The destination checks sign-in and current access. Profile, signed-in sessions and the Assistant library have separate readers; linking to them does not promise a complete export. Original import files may already have been removed after confirmation under their stated policy.

Separate export demonstration

Open the demo explicitly. It uses only eleven fictional records across six categories and copies nothing from your account. Choose categories and inspect their records before downloading JSON or CSV. Amounts remain exact strings. CSV uses one row per original field, standard quoting and a protective apostrophe before text a spreadsheet could interpret as a formula. JSON retains the original text.

Erasure review and demo limits

Erasure review covers the entire fictional account regardless of the earlier export selection. Inspect the effect on sign-ins, sharing and own text, give consent and enter exactly DELETE DEMO. The result empties only the current demo export. Changing the selection or resetting invalidates an old review.

The complete original fictional fixture remains separately downloadable, and reset restarts the demo. This demonstrates the proposed interaction, not physical erasure. Actual export and erasure requests use only the separate actual sections above; the demo neither initiates nor verifies them. Exceptions and copies have their own actual statuses. No demo button erases an actual account.

Multiple statements in one confirmed import

In Portfolio → Import, open Combined broker files (2–6). Add original XML, CSV or XLSX and select the exact supported profile for each file. Every file keeps its own account and asset mappings, time and fee choices. Prepare each conversion, then the combined preview. Switching between files retains unfinished forms.

Choose 2–6 different files, up to 500 kB each, 2 MB in total and 2,000 source records. Review new, duplicate, conflicting and excluded records across all files. Supported stable operation IDs are evaluated together; matching dates or amounts alone do not establish a duplicate. Any conflict or unresolved record blocks the whole bundle. Only separate confirmation writes all new events together.

If a reply is uncertain, explicitly check the original state or retry the exact original operation first. Cancelling a wait does not undo a confirmed write. After success, open saved history and original recipes; supported Undo has its own preview and conditions. Confirmed original files are not archived on the server. Unfinished forms and exact recovery live in open-page memory, not a backup after reload.

Supported file

In Portfolio, open Import → Brokers. Choose Interactive Brokers Activity Flex XML, XTB Cash Operations CSV, Trading 212 History CSV, DEGIRO Account CSV, Fio e-Broker CSV or eToro Account Activity CSV/XLSX up to 500 kB. The first four profiles use UTF-8; for Fio, explicitly select UTF-8 or Windows-1250. eToro reads original XLSX or a UTF-8 CSV of the specific Account Activity worksheet. These are bounded formats and operations; they do not establish full support for every broker statement.

No broker credentials are requested. Select your own file; no demo trade is added to your portfolio. Loading and local review do not write to the journal.

XTB: currency and manually verified costs

For dividend income from a supported Cash Operations CSV, select the XTB History and dividends (v2) scope. First explicitly choose the account currency from the actual statement. Map each dividend to an asset and confirm that its original Amount is gross dividend cash. Amount, currency and gross meaning are not inferred from a comment or ticker.

Manually allocate original negative Withholding tax records with the same account, ticker and exact UTC clock. A tax row can belong to only one dividend and is not counted twice. If withholding is actually zero, explicitly confirm that fact; a missing row does not establish zero. Separately provide the evidenced UTC entitlement instant. Cash posting time does not automatically replace it.

Before confirmation, compare original dividend and tax rows with the one prepared dividend and its net amount. Previously supported trades and funding retain their identities across the profile change. The same choices work in Combined broker files and remain in confirmed history. This does not support interest, unclear FX/Category, unproven NET meanings or native Full Report XLSX.

Save the Cash Operations section as semicolon-separated CSV. The adapter recognizes two exact layouts of 8 or 10 columns with the original account number and UTC window. This does not support native XLSX Full Report, arbitrary CSV exports or changed headers. Encoding is not guessed.

Explicitly choose the source account currency. For every stock trade, provide the actual quoted-price currency, fee currency and amount, and the reason or evidence for this addition. A blank fee is not zero; enter zero only when you have verified it. These are your additions, not facts found in the CSV. The confirmed private recipe retains them.

The first version requires matching currencies and an exact reconciliation of quantity × price and fee with the original cash movement. An unexplained difference, FX or rounding blocks conversion. A Total summary is not an extra transaction and may be explicitly excluded with a reason. Dividends, withholding, internal transfers, CFDs and corporate actions remain unresolved until supported or explicitly excluded.

Trading 212: statement version, time and total costs

Two evidenced older History CSV layouts with 15 or 19 columns and a Time field are supported. Choose the time zone from your original statement; the field alone does not prove UTC. Newer exports with Time (UTC), Gross/Net, CFD and other headers must not automatically be presented as supported.

The file contains no source account number. My Trading 212 statement is a local key; select your own Portfolio account. Explicitly assign instruments. The price and its currency remain from the CSV. For each trade, provide the actual total fee, its currency and the evidence used to verify it. Zero is allowed only as your explicit verified input.

A stated conversion component in the second layout must neither be counted twice nor hidden: it must be included in the verified total fee. The first adapter requires matching price, Total and fee currencies with exact cash reconciliation. Different currencies, an unclear exchange rate, ambiguous IDs and other operations remain unresolved. Dividends and interest are not converted into deposits.

For dividends, select Trading 212 · history and dividends under Trading 212 statement scope. The original v1 trade-history option remains separate. Changing scope clears the loaded file and choices, not a retained unresolved server command. The new option uses the same exact older 15/19-column layouts with Time; it does not support arbitrary newer exports.

For a positive Dividend (Ordinary), explicitly confirm under Dividend evidence that the original Total is gross cash before separate withholding. Its amount is not overwritten. Manually select each corresponding negative Dividend (Dividend tax); the original Time, ISIN and currency must match. Similarity or an offered candidate does not automatically pair records. One withholding record cannot be assigned to two dividends. Explicitly confirm evidenced zero withholding separately; an absent row does not prove zero.

Enter Evidenced entitlement date · UTC and Evidenced entitlement time · UTC from your own records with an explanation. They start blank; the cash-payment time is not copied into them. Entitlement cannot follow payment, and the statement time zone does not shift these explicitly entered UTC values. Conversion creates one dividend event with gross amount and withholding; cash is their net difference. The allocated tax record remains in the audit map without being posted again as another outflow.

Inspect the complete preview, including links and new, duplicate and unresolved records. Back to editing confirms nothing; only the separate confirmation writes to the ledger. Original operation IDs share the namespace used by the v1 trade profile: identical previously stored operations are not added again, while changed content is a conflict. History and its separately confirmed Undo use the events actually added. After confirmation, the original file hash, recipe including dividend links and record map remain, not the original CSV or a restorable copy.

An already net Total, unclear gross amount, missing entitlement evidence, FX and unsupported actions are not guessed. In a supported statement, resolve the record or explicitly exclude it with a reason. New Time (UTC) and Gross/Net headers remain unsupported. This bounded conversion does not establish completeness of an actual broker statement and is not a tax calculation.

DEGIRO: source records, minutes and allocated fees

For income from the supported Account CSV, choose DEGIRO · history and dividends (v2). The profile recognizes the exact original descriptions Dividende and Impôts sur dividende. For each positive dividend, explicitly evidence Change amount as gross cash in its original Change currency, rather than an already net amount. Negative dividend corrections, interest and unclear FX remain unresolved; they do not become inferred deposits.

Select tax records individually from your evidence. Product, ISIN, Change currency, Date, Time and Value date must match; withholding is deducted once and a row cannot belong to two dividends. Without a tax row, separately evidence and confirm zero. Enter entitlement independently in UTC; neither payment time nor Value date supplies it. The same choices remain in combined files and the private history recipe. After confirmation, open the dividend report for that portfolio.

The supported Account CSV has an exact twelve-column header and French decimal-comma cash amounts without thousands separators. Two amount cells have no header of their own, so the adapter preserves their original positions. Transaction CSV, a Portfolio snapshot and native XLSX are different formats. My DEGIRO statement is a local key; this source does not establish a broker account number.

Choose the actual statement time zone and explicitly confirm the number format and use of the recorded minute start for accounting order. The source has HH:mm without seconds; 00 is not an evidenced execution second. Order IDs may be repeated or empty. The wizard uses the original file and row position: it protects exact retries, while changed or overlapping files need your review and explicit exclusion of already posted records. Equal amounts are not automatically merged.

A trade shows its original quantity, price and currency. Verify exact agreement with the cash movement before fees. Supply the evidenced total fee and reason; if it is reported separately, explicitly allocate its original record to the correct trade. Allocated amounts must sum exactly in the same currency and time. A record is included once in the trade and is not posted again separately. Unallocated brokerage commission stays unresolved; exchange connection expenses may be separate fees.

A cash-price difference is not converted into an invented fee or replacement price. Dividends/withholding remain unsupported in the original v1 profile. FX, sweeps, internal flatex transfers and unknown operations stay visibly unresolved or require reasoned exclusion. The adapter does not create missing acquisition history; journal validation blocks a sale without established holdings.

Fio: Czech encoding, source records and net amounts

Upload the original file and choose its actual encoding. SHA-256 identifies the original bytes, not re-encoded text. The wizard displays decoded records and retains the encoding choice in the audit. After changing encoding, select the file again; the wizard does not silently try another decoder.

The exact Czech fourteen-column semicolon Trades header is supported, or an explicitly configured fifteenth Trh column. This does not certify every default Fio export. Preamble records are retained and need reviewed exclusion with a reason. Money pages and balances are not automatically deposit events.

Choose the actual time zone and acknowledge decimal commas and recorded minute starts for accounting order. Settlement date does not replace trade time. Map the account and instrument using symbol, currency and evidenced market. Original-file-and-record identity protects exact retries; changed overlaps need review of the original history and explicit exclusion of already posted records.

Buys and sells must exactly match the original net cash amount in one currency, including fees. A reported fee is preserved; a blank fee needs an evidenced manual value and reason and is never automatically changed to zero. Amounts in different currencies are not added together. Explicit cash-desk transfers additionally require external-flow review with a reason. Unknown operations, FX, dividends and unestablished transfers remain unresolved.

eToro: original XLSX, Amount meaning and actual stocks

Explicitly choose original XLSX or a CSV of the Account Activity worksheet. SHA-256 belongs to the original selected file bytes. XLSX is actually unpacked and its named worksheets, relationships and text/numeric cells are read; financial values are not rounded through ordinary binary numbers. A decoded row in the wizard is not the original binary file. Encrypted files, macros, formulas, external links and unsupported structures are rejected. Legacy binary XLS is not XLSX.

Account Activity has a specific eleven-column header with Units or Units / Contracts. Other worksheets, their rows and preambles remain visible. You prepare only supported Account Activity operations; other records need reviewed exclusion with a reason. Closed Positions does not establish complete opening history, and Summary balances are not deposits.

Choose the actual time zone and account currency, and acknowledge decimal points and weak file-and-record identity. Position ID may repeat across opens, partial closes and fees; it is not an individual execution identity. Equal positions are not merged. Changed or overlapping files need your review of previously posted history.

For a Stocks trade, explicitly attest an actually owned, long, unleveraged position. Supply an evidenced unit price, its currency, aggregate fee and supporting reason. Source quantity remains unchanged. Review Amount meaning separately: gross value must exactly equal quantity × price; on a close, original position capital can be explicitly evidenced without treating it as proceeds. Price and proceeds are not derived from Amount / Units, profit, Realized Equity or Balance. This adapter does not support CFDs or short/leveraged positions.

If an aggregate fee includes a separate SDRT record, explicitly allocate it. Position, time, description and exact sum must agree, producing one cost. Other supported standalone fees require confirmation that they are not included in a trade, plus a reason. A missing fee is never changed to zero. Withdraw Request does not confirm a settled withdrawal; dividends, refunds and unclear cash/FX operations remain unresolved.

Accounts, instruments and original values

Map every source account to a named account in your portfolio. Map each instrument to the correct existing identity or explicitly choose the separate identity offered from its IBKR Conid/ISIN. A similar ticker does not prove the same listing. A new broker identity does not itself provide a market price.

Choose the time zone from your statement settings. An unspecified clock is not converted to assumed UTC. A missing time of day or an ambiguous daylight-saving clock blocks conversion. Review original attributes and the prepared event; an original settlement day is not presented as a verified exact settlement instant.

Unresolved and excluded records

Unsupported or contradictory events stay visible. Correct the choices or explicitly exclude a record with a specific reason. Excluded records are not posted and the resulting portfolio may be incomplete. The reason and original-record-to-row link remain in the private audit.

The adapter does not invent a missing price, withholding or fee. The same date and amount alone do not establish a duplicate. Separate evidenced execution identifiers are preserved; an Order ID is not always such an identifier. DEGIRO here uses explicitly reviewed file-and-record identity.

Preview, confirmation and history

After local review, explicitly prepare the server preview. It compares events with the current journal and separates new, duplicate and conflicting rows. Review the summary and confirm that exact preview. If the journal changed meanwhile, prepare a new preview.

After an uncertain outcome, verify the same account and original request state; do not create a replacement import. Confirmation retains a receipt, history and, where eligible, controlled Undo import. The original file and preview are purged; the confirmed recipe, exclusions, row links and hashes remain private. A hash alone cannot restore the file.

1. Choose exact opening and closing

In Portfolio’s Model tab open Time-weighted return · your valuations. Enter Opening · exact UTC instant and Closing · exact UTC instant as YYYY-MM-DDTHH:mm:ss.sssZ. Closing cannot exceed the verified ledger cutoff. Review recorded flows prepares that exact window; Verify current ledger rechecks the source, not your historical prices.

The interval is (opening, closing]: opening value already includes flows at opening, and closing includes flows exactly at closing. Flow review reads the existing server-owned ledger projection at exact closing once and verifies its binding to the same complete currently recorded ledger. Calculation, detail and export use that retained server projection without reading it again. This does not establish that every record was known then. All recorded flows, exclusions and reversals distinguishes included, opening/prior, later, reversed and internal-transfer records.

2. Enter total values at every boundary

Your total valuations include all investments and cash of the same portfolio in its base currency. Enter opening/closing values after flows at their respective instants and values before/after every actual flow group. Simultaneous flows are grouped, including zero net flows; at most 50 instants and 51 segments. You enter both values: after flow must equal before flow plus recorded net flow.

If a flow occurs exactly at closing, its after value must equal your closing value. Without a closing flow, a final segment reaches closing valuation; with no flows there is one segment. Use nonnegative decimal amounts with a point, up to 24 integer and 18 fractional digits. Zero is an actual input, but zero starting capital makes return unavailable rather than neutral.

3. Review values and explicitly calculate

Review all valuations opens Exact review before calculation with window, currency, revision and every boundary. Review does not calculate return. Calculate TWR from reviewed values rechecks the same account, ledger and membership and uses exactly the reviewed values. Draft edits or cancelling a wait do not recalculate the preceding result or overwrite valuations.

Included foreign-currency flows, securities receipts, missing base amount or valuation mismatches may make return unavailable. Unsupported members are not dropped and a null flow is not zero. The model substitutes no historical FX, cost basis or current price. Correct and complete valuations, including income/costs, and ledger coverage remain your assumptions.

4. Read segments and cumulative TWR

The result uses the unchanged exact-flow-boundary-twr-v1 kernel with 48 significant digits HALF_EVEN. Segments are linked geometrically, not by adding percentages. Whole-period return is not annualized and is neither XIRR, Modified Dietz, a benchmark nor tax result. Cumulative points and the complete table share one calculation; selecting a point or Inspect opens exact valuations and original flow members without recalculation.

Unavailable return remains a gap with a reason, never filled with zero. Long values marked ≈ are shortened only for presentation; exact values remain in detail and files. Contributions and withdrawals are not investment profit, and an actual −100% return is not a neutral segment.

5. Retain the whole conditional model

Whole TWR model · JSON/CSV freshly verifies account/ledger and includes the same complete result, all segments, original financial inputs and all your valuations once. A detail or flow page never narrows the file; authentication/session proofs are excluded. Results and drafts stay only in page memory and may not survive a reload. The model writes no ledger, saves no Snapshot and publishes no financial share.

1. Select actual spot and book depth

Open Market Battle, Actual spot · Binance, Spot pair and Levels on each side (5, 10 or 25). It reuses the original book/trade source; Open original book offers the other view. USDT is a separate quote unit, not USD. This is one spot venue, not an exchange aggregate or futures market.

The visible page periodically reads the shared source; Read now requests another read. Synchronizing, reconnecting, sequence gaps, stale events or an unavailable source do not establish a live book. Failures never automatically replace actual data with an example.

2. Depth means selected standing offers

Bid and ask depth uses two sides with a shared scale for cumulative base-asset quantity. The complete table shows price, original quantity, price × quantity in USDT and same-side cumulative totals. Spread is best ask minus bid. Only captured displayed price levels are counted, not order counts, entire liquidity or a guaranteed execution price.

3. Trades mean observed taker messages

Observed taker trades plots original time × price points, sized by quantity and coloured by initiating side. No trades are created between points. A maker buyer means a taker sell; otherwise it is a taker buy. This derives initiating side from the original flag, not people, holders or net inflows.

Counts and base/USDT quantities concern up to 100 messages observed after connection, with their actual oldest/newest times. This is not a fixed minute or complete market history/volume. Zero records does not establish zero activity. Book and trades have separate clocks and are not an atomic snapshot; they are not combined into a score. Liquidations are not provided; order cancellations are not inferred.

4. Pause the same result for inspection

Pause display freezes the displayed result. Opening a point, Inspect, Source detail or trade pagination also pauses reads of the same result. Complete book levels and All observed trades retain original order, IDs and clocks. Resume and read obtains the current source again; a frozen state is never presented as a continuing live connection.

Source detail separates original status/reason, latest book event, initial REST snapshot, trade capture, source assembly and last successful read. It shows sequences, source addresses and method; publication time is unreported. Original document and exact calculation is the normalized input/result, not an archive of every original REST/WebSocket message or durable historical replay.

5. Whole export and source availability

Download whole JSON/CSV retains the same result, all selected levels/observed trades and one original normalized document. A detail or page never narrows the file. Separate export annotation records updating, paused, offscreen, unverified display state without changing original source status. An old live status alone does not establish export freshness.

If the actual source is unavailable, no fictional prices or trades replace it. Market Battle predicts no price direction, creates no trading signal and places no orders. Source availability alone does not establish permission for further public redistribution.

1. Pin candidates in the shared catalogue

In the US stock catalogue, expand Additional tool: compare individual quotes. The panel automatically reads the same two-to-six selection as historical comparison; a closed panel acquires no additional quotes. Edit members in the shared pinned roster. The SEC catalogue is candidate provenance, not proof of a current quote or issuer identity through CIK.

2. Read original quotes and clocks

Opening checks the original catalogue, reads the exact selected listings and re-admits every same input. The complete board then appears without another confirmation. Each quote retains its original name and symbol, venue, currency, price, quote clock and separate capture. The SEC candidate and native Yahoo listing remain separate identities; similar names do not establish their relationship.

An unavailable member is neither omitted nor replaced by another ticker. If an entire input is missing, the successful subset cannot be applied alone. A missing price or clock within a valid input remains labelled unknown; an actual zero remains zero. Cancel operation retains the previous applied board.

3. Whole selection and source refresh

The public board accepts only the entire validated selection. An unavailable member is never silently omitted. Changing pinned members reads the new selection only in an open panel; Refresh selected quotes repeats the read. A completed failure or explicit cancellation does not trigger automatic retries.

4. Distinguish price, currency and clock

Selection order is not a price ranking. Native currencies, including the case-sensitive GBp unit, remain separate without FX conversion, a common price axis or portfolio total. Quotes are neither live nor necessarily simultaneous. regularMarketTime is distinct from capture, board creation and confirmed publication.

Volume and market cap are original numbers; volume unit and market-cap currency are not independently established. Previous Close and Chart previous Close remain two separate reference fields, not a basis for inferred daily change. The board calculates no return, promises no executable trade and is not a saved historical analysis.

5. Inspect the source and download the whole board

Source detail re-admits the same originals before showing exact identity, SEC candidate, source values/clocks, SHA and retention deadline. Original document is concealed in advanced detail until explicitly opened. Download whole JSON or Download whole CSV retains the same result, the entire original SEC catalogue once and each original listing once. Member inspection never narrows the file; a missing original blocks the whole download.

6. Original verification is not a new quote

Hiding the page or panel interrupts a pending first read, which may resume on return. An applied public board is re-admitted against the same retained originals on return without acquiring newer prices. Withdrawn access conceals values. Cache expiry labels the input’s age; after retention expires, a new read must be explicit. Source availability does not establish a redistribution licence.

1. Select an exact listing and window

Dividends and splits prepares actual events for default AAPL within captured coverage on first opening. From a price or native detail, Open events for this listing opens the exact qualified identity. An unsupported listing is not replaced by another stock; Choose a qualified listing opens the supported-source view; it does not validate the originally unsupported stock.

From UTC day and Through UTC day inclusive are UTC calendar days. The first and last source days are clipped to the exact captured interval; draft and applied summary show the actual start and exclusive end. Prepare full captured window fills the draft without applying a result. Choose Dividends and splits, Dividends only or Splits only.

2. Initial result and selection changes

The first valid selection is applied automatically for the timeline, complete list, Passport and JSON/CSV. Later listing, window or type edits remain drafts until explicitly applied again. Refresh source metadata verifies the source roster and coverage without rewriting an original saved result. An unavailable document is not zero events.

3. Timeline markers do not compare amounts and ratios

Source-event timeline uses separate dividend and split lanes. Horizontal position means original UTC time; marker size does not mean amount, ratio or return. Complete event list can be sorted newest/oldest and paged. Pages do not narrow the whole JSON/CSV; amounts are not summed and event kinds share no monetary axis.

4. Distinguish source values from an unknown calendar

A historical dividend is the provider-reported amount in the reported currency. Its adjustment basis is not independently verified; it is not evidence of original cash paid. A split retains its original numerator, denominator and ratio text. Zero is not a missing value. An event date is not independently established as ex-date, record date, distribution or payment date. Declaration, ex-date, record date, payment, status and corrections remain unreported.

Event Passport or a timeline marker opens the exact identity, original key/time token, amount or ratio, numeric tokens, capture and separate source-body/capture checksums. An event posts no portfolio transaction, changes no Yahoo Close price and establishes no received income, entitlement, tax, future payment or total return.

5. Coverage remains available without events

Coverage and original document remains available for an empty result: exact listing, original interval, whole-document counts, family presence, currency, time zone, capture, ingestion, evidence and policy. Whole coverage receipt shows that complete same input. Zero matches means no event in this document, window and filter; reading the whole document does not establish complete market history. Missing families receive no placeholder event.

Whole export uses the same query and manifest as chart and table under current access/source verification. Even an empty result retains coverage evidence separately from actual events. Local-development authorization is not a public dataset licence; broader source and native acceptance remain separate.

1. Choose an original opening and final closing

In Portfolio first explicitly apply unified history. Under Performance curve from this history select a Fixed opening day and later Final closing day. The curve includes at most 30 closing points within 31 calendar days including opening; every day must belong to the same original history.

Review curve and coverage shows the original opening value, each closing, coverage and unavailable reasons. Review does not yet calculate percentages. Only Calculate Modified Dietz curve creates the curve and every period result.

2. Every point starts on the same day

A point for a given closing means return from the same fixed opening through that closing. It is neither a standalone daily return nor a chain of daily percentages. It uses the unchanged method of Period return from this history, including original time weights for external contributions and withdrawals.

Modified Dietz is an approximate nonannualized return. It is not TWR, CAGR, a benchmark, tax or certified intraday calculation. A deposit itself is not return; value change remains distinct from return. Opening uses the original first day’s cutoff, so a flow already in opening value is not subtracted again.

3. Zero is a result, a gap is unavailability

The chart adds no assumed zero point at opening. An actually calculated zero is valid; a missing percentage remains a gap without interpolation. All daily closing points lists opening/closing values, net external contributions, weighted capital, percentage and the concrete reason. An unavailable point can be selected in the table.

Incomplete values, unsupported currency/flow, unsuitable weighted capital or an opening-basis correction can prevent return calculation. A known portion never substitutes for the complete endpoints required for return. A large journal may require fewer closing periods; the job has shared time and size bounds.

4. Open the exact period and all its flows

Select a point or closing day. This point’s period uses the same already-calculated period without recalculation. It shows return, value change, external contributions and weighted capital, then every original flow/classification: included, already in opening, after closing, reversed at closing or internal transfer. A flow exactly at closing has zero time weight.

Flow detail retains the original event, record, reversals and corrections for that closing. Original opening values and Original closing values open shared source detail for the exact day. Flow pagination does not exclude other flows from the period.

5. Whole file and current verification

Whole performance curve JSON or CSV contains every compact point, recipe, calculation identity, methodology and unchanged original history with sources once. Neither originals nor full flow rosters are copied thirty times. Selected point and flow page do not narrow the file.

Verify the same curve rechecks that history and current access without substituting newer prices. A changed draft is not automatically recalculated and the earlier curve does not belong to it. The curve uses currently recorded history and original reference prices; it proves neither a complete investment history nor prices known then. It creates no account save, financial Snapshot or share.

1. Open comparison mode

In Energy switch to Compare zones. Two available registered zones and a bounded common period are prepared; apply the comparison control for your own selection of two or three zones. An unavailable zone is not substituted. Each zone’s native Data Passport and original single-zone view remain separate.

The comparison uses original wholesale auction prices in EUR/MWh at PT15M resolution. These are not household tariffs, real-time executions, forecasts or a new spread calculation. Each series retains its own source and capture time; one result does not imply simultaneous source capture.

2. Review common whole days

The reference timezone is Europe/Prague. First delivery day and Last delivery day select at most seven whole reference-calendar days inside every zone’s captured overlap. Prepare common window fills only a draft; review common whole days and exact prepared UTC bounds. A comparison cannot be applied without a verified common whole day.

A time overlap does not prove every price inside it is available. Missing prices remain gaps. Each source retains its original timezone and offset; the reference axis never rewrites them. Clock changes can alter day length.

3. Apply one common query

Only Apply comparison creates one common query for all selected series. Chart, table, summary, Passport and whole export share that result and manifest. Later draft edits leave the previously applied zone set and dates clearly labelled below until a new selection is applied.

Refresh catalogue and coverage checks registration and time coverage, not new prices. A failed check retains the draft. Neither query nor comparison adds another source, currency conversion, interpolation or replacement zero.

4. Inspect series, intervals and provenance

Original interval prices · common scale gives each zone its own colour and legend. Each horizontal segment belongs only to its original delivery interval. Zero and negative prices are values; a missing or unusable point is never bridged. Future delivery remains an established auction price, not an observed future trade.

Selected original interval and zone opens the exact series and observation through Original interval Passport. All intervals of every selected zone shows the same cohort in a complete paginated table, with reference/original local and UTC times, quality and missingness. No artificial row is created for another zone’s absent counterpart. Value Passport keeps original token, offset, source generation, capture and unknown publication distinct.

5. Export retains the whole comparison

Export analysis reviews the same displayed result and current access to every selected source before JSON or CSV download. Neither table page nor selected point narrows the export. The file retains original series, query, manifest and evidence; it fetches no newer replacement data and creates neither a saved Snapshot nor new redistribution permission.

1. A zone is a delivery market, not an entire country

In Energy open Day-ahead electricity prices. Bidding zone offers only available registered sources; an energy_zone record has its exact code and Data Passport. A country with energy indicators is a different identity. Native zone detail fixes that zone; Other bidding zones links open other actually registered zones.

You are reading a wholesale auction price in EUR/MWh, not a household tariff, a real-time execution or forecast. The previous supported reader remains preserved; the shared view adds exact catalogue identity, one query and provenance.

2. Prepare local dates and explicitly apply

First delivery day and Last delivery day are local calendar dates in the selected timezone, at most seven days inside actual captured coverage. Prepare captured window fills a draft only. Check the timezone and displayed UTC bounds; a local day need not last 24 hours across clock changes.

Only Apply zone and period creates a new price query for chart, table, Passport and export. Editing a draft does not overwrite the previously applied zone or result; its applied period remains above it. Refresh catalogue and coverage checks registration and available dates without creating prices.

3. Each step belongs to its own quarter hour

Each chart step holds one auction price over its own delivery interval. Selected original interval and Original interval Passport connect it to exact provenance. Negative prices and genuine zero are values; a missing or unusable price remains a gap. Quarter hours are not silently converted to daily closes or household bills.

Upcoming delivery · auction price means future delivery at an established auction price, not an observed future spot trade. Delivering-now and delivered intervals remain distinct. The displayed interval end is exclusive; original canonical period_end is one microsecond earlier, derived from documented PT15M rather than the next row.

4. Whole table and exact local offsets

All original intervals retains every row of the selected result, its price or missing reason, delivery state, original UTC bounds and quality. Local offsets distinguish repeated hours. A page holds up to 96 intervals, not a fixed day length; pagination narrows neither chart nor whole export. A row’s Value Passport opens the same shared detail as the chart.

5. Distinguish generation, capture and publication

Value Passport preserves the original offset timestamp, numeric token including zero or null, unit, zone, resolution, requested local dates and original licence statement. Document generated at source is not a proven auction publication time or Datisto’s first capture. Observed by Datisto, input storage and query computation are distinct clocks; unknown publication is not estimated.

6. Export the same result and handle unavailability

Export analysis reviews the displayed result and current access before Download JSON or Download CSV. The whole result, query, method, manifest and original evidence share the chart/table scope; the selected page does not narrow it. JSON retains complete typed information; CSV protects text cells against formulas. Download creates neither a saved Snapshot nor new redistribution permission.

Missing registration, coverage, source or current verification stays explicit. This view substitutes no other zone, newer original or replacement price. Refreshing the catalogue or the same result is not an electricity order and unlocks no unrelated private data.

1. Enter a name or symbol

Enter at least two characters of a name or symbol in Asset, model, area or article… in the header. Public data catalogue loads after a short typing pause; an invalid or oversized query is not sent to the catalogue. Catalogue search accepts 2–120 characters.

Up/down arrows select a result, Enter opens it and Escape closes the dropdown. Clear search removes the query. Other search branches retain their own group labels.

2. Distinguish actual types and status

Catalogue · … groups distinguish company, stock listing, AI model version, AI model offer, country and region using each record’s actual type. A hit shows its exact identity, optional symbol and Active, Planned or Retired status. A model offer is not a model version; neither a country nor an Energy section link is an energy zone.

The catalogue need not contain every object family or the whole market. An empty search does not mean a zero price or prove the object absent outside this catalogue. An exact known identity already shown as an actual market card may be omitted as a duplicate; a matching name or ticker does not prove identity.

3. Open the appropriate detail

Open Data Passport opens the exact catalogue object. AI offers use Open model offer and model/version records use Open model detail; supported AI identities may redirect to their native AI view. Planned and retired records use View record. Catalogue status does not establish a current price, an available time series or source-data permission; the detail checks its own conditions.

Nonfinancial objects receive no invented stock quote or benchmark. AI offers · OpenRouter and Working articles · not approved remain distinct from catalogue types and approved publication.

4. Catalogue failure preserves the query

Public data catalogue has its own loading, no-match and failure states. Retry catalogue repeats the current query; a failed branch does not relabel other results as its replacement. A changed query or language invalidates older responses so they cannot replace newer search results.

5. Public metadata is not a private index or permission

This branch reads bounded public catalogue metadata without account cookies. It does not index private portfolio names, watchlists, notes, results or credentials. It creates no new provider index, downloads no prices for search and grants no rights to further data.

1. Select two original days

In Portfolio first load, review and explicitly apply unified history. Under Explain the value change select an Opening day and a later Closing day. Both must belong to the applied history; this panel acquires no new prices.

Review original values shows both whole values, known portions, valued holding/cash counts and original day boundaries. Review alone does not calculate the breakdown. Only Calculate value-change breakdown creates a result.

2. Read numerical contributions in their defined order

The bridge follows quantity → native price → FX rate. Cash-balance change is separate; its currency movement belongs to FX. Interaction terms are already assigned to the following price or FX component and are never added twice.

The chart shows every contribution, including Unattributed residual and Rounding difference. The sum matches the compared change only when the result explicitly confirms reconciliation. Exact amounts and their origin remain in the table and detail; the chart is not a separate calculation. Any difference between the total change and the sum of individually rounded row changes is disclosed separately above the table; it is not added again to chart contributions.

3. Distinguish whole values from known portions

If either endpoint is incompletely valued, the result is named Known-portions comparison only. Check both endpoint coverages, opening-only/closing-only members and unavailable members. The difference can include coverage changes; it is not a complete portfolio change or return.

A new or closed holding, changed source, native unit or unit basis may prevent attribution. An available difference remains in the residual with a reason. Proven absence may establish zero held value, never zero price. Missing prices or FX are not filled; USDT and BTC are not USD.

4. Inspect all members and their origins

All holdings and cash balances contains every holding and cash account in the comparison. Pagination changes presentation only. Select a member or endpoint value and use the single shared detail to inspect original quantities, prices, units, fixings, source identities and any residual reason.

Original opening sources and Original closing sources open the existing source detail for that exact day and member. Matching tickers do not prove identity.

5. Whole breakdown and the same history

Whole value-change breakdown JSON and CSV retain all members, contributions, coverage, methodology and the unchanged original history with sources once. The current page and open detail do not narrow the file. Download verifies the same originals and current access, never substitutes newer prices.

When the same input cannot be currently verified, the breakdown is not displayed as valid. Verify the same breakdown rechecks original inputs; a retained draft or earlier file does not override current access or source conditions.

6. Value change is not investment return

Deposits, withdrawals, transfers and quantity changes are not themselves gains. This numerical bridge is not TWR, Modified Dietz, market causation or complete dividend/fee attribution. For a separate approximate return use Period return from this history with its own cash-flow review; missing inputs remain unavailable.

1. Name the applied history configuration

In Portfolio first prepare and explicitly apply the historical report. In Save applied history name the actual applied configuration: your portfolio, fixed UTC interval, explicit source choices and presentation. New unfinished source choices are not saved automatically.

This saves settings for fresh work, not prices, quantities, returns, original source files or a financial Snapshot. Start and end dates stay fixed; reopening later does not automatically move them to a current interval.

2. Review and confirm the save

Choose Review history save and inspect the name, portfolio, full interval and exact source families. Only Confirm save to account saves the configuration and returns an Immutable command receipt. Review is neither a new financial calculation nor a journal change.

If another tab changes the collection, the draft does not silently overwrite its newer version. Verify the current collection and review your retained draft first. An oversized configuration is not automatically truncated.

3. Return through your private library

Find the same named configurations in Portfolio, My analyses and Dashboard return to work. The list requires current verification of your account; after expiry its names and settings stay concealed until verification is renewed.

Rename and remove also have their own review and confirmation. Removing a configuration does not change the portfolio, recorded trades or original financial files. If its original portfolio is no longer available, the saved link does not select another one.

4. Opening prepares a draft only

After opening, Review saved history rechecks the same private configuration and shows its conditions for adoption. Adopt as draft only does not download sources or calculate a report. Overwriting existing work requires explicitly choosing Replace draft with saved settings. An unresolved retained request must be resolved first.

Adoption sets the draft interval and logical source choices. A previous unit confirmation is your saved assertion, not fresh source verification. A changed session or late reply cannot adopt another account’s or an old draft.

5. Load and review actual inputs again

Explicitly load the current owned journal and daily states. Review every holding in the interval, including those closed today. Newly present identities require an explicit choice; a missing saved holding must not disappear without explanation.

Stock listings and Coinbase series remain exact logical selections. A Binance pair is resolved again by the same native symbol, base asset and USDT/BTC quote in the current market. An old market capture or similar ticker is not a substitute. A missing, ambiguous or changed source requires your review.

6. Explicitly run valuation and period return again

Only Load selected original histories, review of actual units and Apply unified history produce a new financial report. New sources or journal corrections can change the result. A fixed older interval may fall outside the currently available rolling price window; today’s price does not fill it.

Any saved return dates restore only an interval selection. After fresh valuation, Review period and external flows and Calculate period return are required again. Neither the return percentage nor old price authority is stored with the configuration.

7. Retry an uncertain confirmation exactly

If a connection ends before the save outcome is confirmed, the retained request may already have been accepted. Verify the account and use that same retained confirmation; do not create another copy merely because the reply did not arrive. Opening a page does not automatically send a retained request.

A configuration helps you return to your own work. It is not a financial Snapshot, a full portfolio backup, sharing with other people, a trade instruction or proof of then-known information or complete investment history.

8. Save the whole original result separately

To retain the exact original numbers, choose Save original result beside an actually calculated current or unified historical valuation. Review the portfolio, interval, inputs, currency mode and partial coverage. Separate confirmation saves the whole financial Snapshot in My analyses; named history settings remain a different feature.

Open the original valuation from My analyses and verify access to that same result. Today’s prices or later transactions do not replace its original holdings, cash, daily points, unpriced members or inputs. Whole JSON and CSV belong to this original, including all table pages. Optional separately calculated returns or allocation are not automatically part of the valuation.

Use conditions in a new draft transfers only the portfolio, interval and source, unit and currency choices. Confirm preparation, verify the current owned portfolio, then explicitly consent to adoption or replacement of existing work. Prices, quantities and financial results do not transfer; you explicitly start any subsequent acquisition and calculation.

Saving requires the exact original inputs to remain available. A missing archive or withdrawn current access is shown as unavailable; no substitute price fills it. Your unfinished result remains retained. The saved valuation is private and is neither a trade instruction, a portfolio backup nor confirmation of complete investment history.

1. Select two original days from the same history

In Portfolio first prepare and explicitly apply Unified history. In Period return from this history choose an Opening day and a later Closing day from that same report. Values are not entered manually and today’s price is not added to the past.

Both endpoints need complete nonnegative values in the portfolio’s actual base currency. A known portion is not the whole portfolio. Missing USDT/BTC conversion, a missing price or an overly wide original amount is not rounded or inferred to make a return available.

2. Review the whole period and external flows

Choose Review period and external flows. The review shows original opening and closing values, coverage and all relevant recorded deposits, withdrawals, security receipts and own transfers with inclusion or exclusion reasons. Reviewing or changing a date does not calculate a new return.

Opening value already includes the opening day’s cutoff. A flow exactly at that instant is therefore not subtracted again. Only flows after opening through closing inclusive are included; a flow exactly at closing has zero time weight.

3. Distinguish contributions, transfers and corrections

A deposit is a positive external contribution and a withdrawal a negative one. A transfer between own accounts does not increase external capital. Flow membership belongs to the selected closing day, not automatically the last day of the wider history; a later reversal does not rewrite an earlier close.

If a reversal inside the period cancels an event from opening or earlier, it may change the basis of opening value itself. This first model marks that correction unsupported and retains the original references. It does not invent a withdrawal or turn an accounting correction into investment loss.

4. Explicitly calculate the approximate return

After review choose Calculate period return. Modified Dietz compares value change with net external contributions and their time weights. It displays value change, net contributions, return numerator and weighted capital so the percentage can be assessed.

This is an approximate nonannualized return for the selected period. Zero contributions are not missing data; a missing historical amount is not zero. Nonpositive weighted capital and unsupported included flows have specific unavailability reasons.

5. Inspect original flows and their weights

The table and shared flow detail use the same result. Review the exact original amount, unit, economic instant, record and related reversals. Time weight uses the actual distance to period end; its original time numerator and denominator remain inspectable.

The first model accepts only known consistent cash flows in exactly the base currency. A foreign-currency deposit needs its own historical value, and a security receipt cannot be valued using cost basis or today’s price. An opening or closing fixing does not repair those gaps.

6. Download the whole result from the same inputs

Whole period return JSON or CSV preserves both original endpoints, every included and excluded flow, weights, reasons and result. The complete original financial history is attached once. Selecting a flow, page or day does not truncate the export.

Changing days or inputs requires fresh explicit review and calculation; a late old reply cannot replace them. Display and download require current access to the same ledger and sources. A file preserves provenance and methodology, not a login or live authority.

7. A reference is not exact TWR or then-known knowledge

The model uses original end-of-day references, which may not coincide with a deposit or withdrawal. Calculation uses 48 significant digits with disclosed numerical rounding; the result does not claim arbitrary mathematical exactness.

TWR would need valuations immediately before and after every external flow. This view does not claim TWR, CAGR, a benchmark comparison, tax or complete corporate actions. Later-recorded corrections and histories captured now do not prove what the application knew then.

1. Prepare actual historical holdings

In Portfolio open Historical value and keep Unified history. Load completed UTC days; the roster includes positions closed today. Today’s quantity is not projected into the past.

Each holding and accounting unit has one source family. Stock, an original Coinbase series, a selected Binance pair and No source cannot overlap for the same member. Adding a pair does not rewrite original stock or Coinbase prices.

2. Review the whole original pair catalogue

Choose Load current Binance market. For a particular own holding select Binance · select exact pair and use Search original Binance pair to find the exact symbol. Review the base asset, quote unit and original catalogue input, then Select pair for that member. Its count belongs to a particular capture; it is neither downloaded history for every crypto nor a guarantee of current listing.

Available choices are admitted spot pairs quoted in USDT or BTC, including original names in other scripts. This does not add USDC. A matching ticker does not prove identity: confirm that one own unit means one unit of the selected pair’s base asset.

3. Load originals and apply the whole draft once

Review all source families, exact pairs and unit confirmations. Load selected original histories obtains only the selected unique sources, at most eight pairs; it does not download the whole catalogue or issue a separate request per day. Loading and draft edits do not calculate a report.

After reviewing the original price inputs choose Apply unified history. One result supplies the shared chart, complete table and inspector. New pair assignments stay in private working memory; they are not automatically saved into older price mappings and do not change the journal.

4. Read the original day and all time limits

Quantity and cash remain at the last millisecond of the UTC day. A Binance candle belongs to that exact day and its original close time plus one millisecond is the following midnight. This separate price boundary is neither the candle start nor an equity exchange close.

Only the completed matching day from the originally captured rolling window of up to 90 days is used. A missing day has no older replacement price. The unfinished candle stays in the original input but does not value the portfolio. Binance requires positive original OHLC; another source family’s zero rule is not transferred here.

5. USDT and BTC are separate native units

Value is the exact dated quantity times original close in its actual quote unit. Switch the chart and totals to USDT or BTC without mixing different units on one axis. Cash is counted exactly once. Cost PnL requires known cost in exactly the same unit.

USDT is not USD and BTC is not a dollar. Without a separately evidenced historical bridge their base conversion remains unavailable. CNB converts only actual supported fiat units; the known portion is not the whole total. Native pair value remains available even when base conversion has a gap.

6. Verify the same source and download the whole interval

Verify the same unified history checks the same financial result, original catalogue and every selected history without replacing prices. An expired or removed original requires a new explicit load and review. A late old reply cannot overwrite the new draft.

Whole unified history JSON or CSV preserves all days and members, the original unified report once, the whole original catalogue once and every complete selected price input once, including all original fields and excluded candles. Selecting a day or unit does not shorten the file. Login and live authority do not travel with it.

7. Reference value is not return or then-known knowledge

Source provenance, capture and exclusions are inspectable in member detail. A specific spot pair’s price is not a global aggregate price, executable offer or proof of your asset mapping. History acquired later does not prove what the application knew then.

Deposits and value changes are not automatically investment profit. This view does not calculate TWR, CAGR, performance drawdown, tax or historical FX PnL and does not add unrecorded corporate actions.

1. Prepare the interval and actual dated holdings

In Portfolio, prepare a completed UTC interval in Historical value. Unified history uses actual daily states from the current recorded ledger, including holdings closed today. It does not project today’s quantities into the past.

Unified history leads to one combined result. Original stock view remains separately available; adding crypto does not rewrite its prices, quantities or files.

2. Confirm a source for every historical holding

For each own holding and accounting unit, explicitly choose No source, Confirmed stock listing or an available daily Bitcoin, Ethereum or Solana series from Coinbase in USD. Confirm the unit where requested. Matching ticker text or an old current-price assignment does not prove the correct historical source.

Adopt confirmed stock sources copies reviewed stock choices into the draft. A missing or unassigned holding remains in the result without an estimated price. Selection is neither a trade instruction nor an automatic change to the saved ledger.

3. Load original inputs and apply the complete draft

Optionally enable Use daily CNB fixings to convert into the actual CZK, EUR or USD base currency. Then choose Load selected original histories and review every returned source, unit and time window. Loading inputs or editing choices does not yet produce a new financial result.

Apply unified history calculates one report from the complete reviewed draft. Stock decisions remain original; crypto values use specific original daily closes. A current-price overview is not treated as daily history.

4. Read the two actual time boundaries

Quantities and cash belong to the final millisecond of the accounting UTC day. The daily crypto candle completes at the following UTC midnight; the detail shows that separate price boundary and the original candle start/end. A candle’s start is not its closing time.

Only the exact matching completed UTC day is used. A missing day, null, suspect, retracted or invalid candle remains a gap without carrying an older price. A valid original close may be zero. Each source has its own capture clock; downloading history later does not prove then-known information.

5. Inspect native and base units

Read the chart in native units or the base currency. The complete table and shared day/member/source detail use the same result. Every holding and cash balance is counted exactly once; the known portion stays distinct from the complete portfolio, and an unknown value is not zero.

Conversion requires the needed original rates; a member already in the base currency needs no unrelated fixing. A CNB reference may carry a real earlier fixing for up to seven calendar days, with visible age and no bypassing the latest unusable point. GBP/GBp or USDT are not automatically equated with another unit. The local CNB dataset is currently scoped to 2026, with the lead-in bounded by 1 January. Outside that interval, turn off currency conversion to continue in native units.

6. Verify the same inputs and download every day

Verify the same unified history renews access to the same report and original inputs without silently recalculating with different prices. A changed source draft, ledger or session cannot adopt a late old result. Temporarily concealed private work can return after fresh admission.

Whole unified history JSON or CSV contains the full interval, every member, the original financial report and complete source artifacts. Selecting a unit or day does not truncate the export. A file preserves provenance and methodology, not a login or live authority.

7. Value and cost PnL are not performance

Crypto value uses original quantity and close, while currency totals use exact original amounts before final rounding. Cost PnL requires a known cost basis in the same original unit; it must not be confused with base-currency PnL.

Deposits and withdrawals change value; they are not investment profit. This reference does not claim TWR, CAGR, performance drawdown, complete corporate actions, tax or historical FX PnL. Reconstructing the current recorded ledger is not a record of what the application knew in the past.

1. Prepare the original history first

In Portfolio, open Historical value and switch from Unified history to Original stock view. Prepare completed UTC days and review the actual dated positions, cash and assigned stock sources, including positions closed today. Conversion uses that exact result without changing its quantities, prices or original export.

The base currency comes from your portfolio. This workflow supports exact CZK, EUR and USD units; it does not infer a rate for GBP, GBp, USDT or other unsupported units. The local FX dataset is currently scoped to 2026. Up to seven days before the interval are loaded, but never earlier than 1 January 2026; a missing older reference remains a gap. An interval outside the supported year disables loading with an explanation.

2. Load and review both rate histories

Choose Load original daily fixings. This prepares two separate original inputs: CZK per 1 USD and CZK per 1 EUR. The window covers the selected interval and up to seven preceding days for explicitly carried references. Loading inputs does not yet calculate the financial conversion.

Review the units, returned time window and both original results. Then choose Apply historical conversion. A single current fixing is not applied to the whole past interval.

3. Distinguish the complete total from the known portion

Historical value in base currency uses the original value of every dated position and cash balance exactly once. A complete total is available only when every member is covered; otherwise the known portion is labelled and each unavailable part has a reason. An unknown value is not zero.

A member already in the base currency needs no unrelated rate. Conversions and totals use exact original values; displayed amounts are rounded independently, so summing them may differ from the total in the final decimal places.

4. Open a day and its original fixing

The chart and complete date table use the same result. Selecting a day from either opens the same detail with positions, cash, gap reasons and rates. The original fixing date and its age stay distinct from the valuation day.

The latest original fixing no later than the day is used, with an age limit of seven days. A weekend may carry a real earlier reference; it does not create a new weekend observation. If the latest point is null, suspect, retracted or nonpositive, that conversion stays unavailable rather than silently falling back to an older usable value.

5. Verify the same inputs and download the whole result

Verify the same historical conversion renews access to the same original result and inputs. A new session, changed ledger or revoked access cannot expose old financial values without fresh admission. A late reply cannot overwrite an edited draft.

Whole base-currency history JSON or CSV contains the full interval, every member, the original financial history and both complete FX inputs. Selecting a day does not shorten the export. A file preserves methodology and provenance, not a login session or live authorization.

6. Value is not return or then-known knowledge

The curve reconstructs value using the current recorded ledger and sources retained now. A later-recorded event or newly captured historical reference does not prove what the application knew then. A fixing is a reference rate, not an immediately tradable quote.

Deposits are not profit, and a change in value is not TWR, CAGR or performance drawdown. Conversion also does not determine historical FX PnL, tax or missing crypto history. Those methods need their own suitable inputs and remain separate.

1. Choose an interval and load daily ledger states

In Portfolio open Historical portfolio value and switch from Unified history to Original stock view. Enter From and Through inclusive, at most 90 completed UTC days. Load daily ledger states reads one owned batch of daily projections from the same complete journal; it is not 90 separate price reads. Each cutoff is 23:59:59.999 UTC, not an exchange closing-trade timestamp.

Today’s or a future incomplete UTC day is not used. Calculation follows recorded events effective at each cutoff, including dated cash and positions already closed today. It introduces neither another browser ledger nor today’s quantities in the past.

2. Review sources for closed positions too

From the roster held during the interval, include assets you no longer hold today. Use the exact catalogue candidate, explicitly read history and review the separate native listing, type, currency and one-share unit. Your unit confirmation is not independent proof of a SEC issuer-to-price-provider join.

The source is a fixed original 3mo history, with at most eight distinct sources. Its actual coverage may differ from the chosen interval; dates before source coverage remain gaps. This initial view does not supply historical crypto or FX by borrowing a current price or today’s fixing.

3. Explicitly apply historical valuation

After review choose Apply historical valuation. Each day uses actual quantities and cash from its daily projection, the latest eligible completed original Close and its exact time rule. Sources, interval and ledger bind one report; a draft change does not silently recalculate it.

A price may be at most seven days old. Completion requires the relevant source-reported session end or conservatively timestamp plus 24 hours; this is not a verified historical close clock. A latest null is not bypassed with an older number. A weekend may retain an eligible earlier real point with explicit age, but no weekend observation is fabricated. An unresolved recorded split-basis change remains a gap.

4. Read one unit and honest coverage

Use the native-unit selector to choose the chart’s currency or unit; one axis does not mix USD, GBP, GBp or other tokens. A known-portion curve is not a full value. A complete base total exists only on days when every valued position and cash balance exactly matches the portfolio base unit; otherwise it has a gap. All daily states retain other native totals and coverage reasons.

Value changes are not automatically profit. A deposit is a contribution, not investment return; event markers are actual records from that day, not automatically external cash flows alone. The view does not calculate TWR, CAGR, performance drawdown or historical FX PnL.

5. Inspect a dated state and original point

Open Day detail from the chart or complete table. Review dated positions, cash accounts, actual events, exact selected price/unit, age and gap reason. Original Close and separately reported Adj Close are not an application split adjustment or proof of dividend total return.

This is the current recorded ledger reconstructed at an economic cutoff. Later recorded corrections and later captured sources may contribute to today’s reconstruction; the report therefore does not claim what was known then or complete corporate-action coverage.

6. Export the whole interval, not just a chart point

Whole value history JSON/CSV retains original economic inputs once, all daily projections, holdings, cash, prices, gaps, totals and the manifest. A selected day, native unit or page does not narrow the file. Current owner/session, the same ledger and every same source original are rechecked before export.

Changing the interval, mapping, ledger or admission requires explicit renewed review. The file contains private financial inputs, conveys no current access authority and is neither a canonical Snapshot nor public sharing.

1. Prepare allocation from the same valuation

In Portfolio, first review sources and calculate Value in portfolio base currency. Then open Allocation including cash and choose Prepare allocation. It retains the same original report, prices and fixing day; this is neither a new price read nor another quantity-times-price calculation.

Changing the original report, ledger, source admission or session invalidates its dependent result. An invalid constituent is not silently replaced with a newer value.

2. Inspect complete groups and original members

Securities group by the exact owned asset ID across accounting rows; cash groups by original currency across accounts. Two different assets remain separate even if they use the same listing. Group detail shows all member positions or cash accounts and original price/rate inputs.

The compact chart shows at most the eight largest groups and states their count against the full report. All others, including unknown groups, remain in the complete table; no group is omitted. This is direct reported exposure in the base currency, not fund look-through, sectors or complete economic currency risk.

3. Distinguish the whole portfolio from its known portion

With incomplete coverage, read shares only as the known portion and inspect uncovered members. A missing value is not zero. A negative original member remains visible even when a positive member makes its group net positive; an ordinary chart or plan must not conceal that state through group netting.

A zero or nonpositive denominator and precision-budget failure have explicit unavailable reasons. Groups, shares and targets use exact original conversion fractions, not displayed rounded amounts. Every output reports its own possible HALF_EVEN rounding to at most 18 decimal places.

4. Prepare, load and save your own target weights

In Target weights enter an exact percentage for every current group totalling 100%. Planning supports at most 50 complete groups; a larger report is not truncated to its first 50. Reconcile missing or extra groups explicitly, without automatically equal-weighting a remainder.

Load saved targets into draft only adopts reviewed owned settings. Review saving targets shows differences; only Confirm and save targets writes a new revision for the same current ledger and groups. Clear saved targets requires review and writes an empty revision; it does not erase draft targets or the ledger. Loading, saving or clearing neither reads prices nor calculates a plan.

If the saved revision or ledger changes, prepare a new review. After an uncertain write use only Retry exact original Save; a new draft does not rewrite the original request.

5. Explicitly review target allocation

Review target allocation requires complete usable coverage, a positive exact total, no negative original member and complete targets totalling 100%. It shows current/target shares, percentage-point drift, target amounts and hypothetical gross exposure adjustments in the same base currency.

Exact unrounded adjustments cancel; independently rounded displayed rows need not sum to the string zero. These are not trade quantities, execution, fees, taxes or ledger changes.

6. Export the whole report and original inputs

Whole JSON/CSV retains the original valuation once, every group/member, coverage, own targets and any plan or unavailable reason. A chart, page or open member does not narrow the file. Export rechecks owner, session, the same ledger and every same original source.

A private export is neither a canonical financial Snapshot nor public sharing. It does not prove historical returns, complete corporate actions or public data rights.

1. Prepare the original mixed valuation

In Portfolio, first prepare One portfolio valuation: explicitly choose source families, review units and calculate the native report. Then open Value in portfolio base currency. Conversion neither substitutes a newer price nor recalculates quantities; the native report remains separate and unchanged.

The actual base currency of your portfolio is used, limited to CZK, EUR or USD. An unsupported base is not switched to USD and the ledger setting is not changed.

2. Choose a day and review original fixings

In CNB fixing date choose a specific day within the permitted seven-day window relative to valuation cutoff. Load original fixings reads the exact source series for 1 USD and 1 EUR in CZK. Review the day, value, unit quantity, quality and availability of both original rows.

An absent, invalid or nonpositive rate is not replaced by zero or a previous day. Weekends do not silently inherit Friday’s fixing. Fixing day, price time, capture and portfolio cutoff are separate facts, not simultaneously tradable quotes.

3. Explicitly calculate the base-currency report

After review choose Calculate in base currency; the button confirms the reviewed inputs without another checkbox. Calculation rechecks current owner/session, the same ledger, original prices and fixing admission. A changed or expired constituent does not silently replace one source.

A value already in the base currency needs no conversion. It stays covered without an unrelated rate; an entirely base-native portfolio needs no fixing. Cross-currency conversion requires only the legs actually used.

4. Distinguish coverage, known portion and full total

The table retains all positions and cash exactly once, original unit/value, converted amount and gap reason. Conversion supports only CZK/EUR/USD. GBP, GBp, GBX, stablecoins and other tokens are not automatically relabelled or rescaled.

The known portion sums covered values. A full total appears only when every position and cash balance is covered; missing exposure is not zero. Totals come from exact original fractions, not a sum of rounded rows. Each output separately reports any HALF_EVEN rounding to at most 18 decimal places or a budget limit.

5. Inspect sources and download the whole conversion

Point detail retains the original price and fixing, exact identities, observations, evidence and clocks. Whole converted valuation JSON/CSV includes one full input, every position, cash, rates, gaps and totals; pagination or a selected point does not narrow the file. Export rechecks the same originals and current ownership/rights.

Value is converted, not historical cost basis or native PnL. The result is not base-currency profit, historical return, tax analysis, a financial Snapshot or public sharing. Later source capture does not prove it was known at valuation cutoff.

1. Open one valuation of your own portfolio

In Portfolio, open Valuation → One portfolio valuation. Inspect your owned positions and the report’s economic cutoff. For each position choose Valuation source: No source, Data series or Stock listing. A family is not selected merely by ticker or by whichever price is newer.

Saved mappings from both families can be loaded into a draft, but they neither choose the position’s family nor calculate anything. If two mappings exist, choose explicitly. No source is a valid choice with an uncovered value, not a zero price.

2. Review the chosen data series

For Data series choose Load available sources, select an actually offered series and inspect its readable name, identity and unit per recorded unit. Explicitly confirm the unit matches your position. Selecting an asset or name alone does not verify that match.

A crypto reference time or daily bucket is not an exchange closing trade. The source retains its own age, quality, admissible-point and rights rules; the stock Close rule is not automatically applied to it. An unavailable source is not replaced by a similar series.

3. Review the stock listing and unit

For Stock listing use the exact current catalogue CIK/ticker/exchange, explicitly load history and review the separate native identity, type and currency. Confirm that the position unit corresponds to one share of that listing and apply the assignment. This does not independently prove the SEC issuer-to-Yahoo-price join.

A stock point must be completed and eligible at cutoff: reported regular-session end or conservatively timestamp plus 24 hours, at most seven days old. A latest null is not bypassed with an older price, and a recorded later split-basis change remains a gap. These are application rules, not proof of a historical closing trade or a complete corporate-action feed.

4. Calculate the whole valuation with one button

After reviewing all selected families choose Calculate whole valuation. Calculation rechecks current owner, session, exact ledger basis and source admission. Do not run two separate user calculations or sum their cash for the combined report. Earlier standalone reports remain a separate advanced option.

One report retains every owned position, including No source and unavailable values. Exact quantity multiplied by an admissible unit price gives value; cash is copied from the original ledger only once. A family, source-draft or economic-vector change invalidates the previous result.

5. Read totals and PnL in the correct unit

Summary values and the complete table separate coverage, original cash and subtotals by exact native unit. USD, GBP and GBp remain distinct, with no FX or 100:1 scaling. A base total is available only with complete coverage and exact agreement of every price/cash unit with the base currency.

Cost PnL requires a compatible accounting unit and supported cost basis. A missing price or incompatible cost is neither zero nor an estimated gain. This report is not historical total return, TWR/XIRR, dividend or tax analysis.

6. Inspect original points and export the whole report

Selected-row detail distinguishes the family, exact original point, price/unit, time rule and source admission. Source timestamp, capture and portfolio cutoff are different facts; a combined report does not make the quotes simultaneously executable or atomic.

Whole valuation JSON/CSV retains original inputs from both families, every position, cash once, subtotals, gaps and one result manifest. A page or point inspector does not narrow export. Before download, current owner/session, the same ledger and every same original are checked again; an invalid constituent blocks the file instead of substituting a newer price.

The file contains private portfolio material; it is not an automatic financial Snapshot, public share or new data permission. Loading or saving existing metadata does not change those conditions.

1. Open your portfolio and valuation cutoff

Sign in, open your own portfolio and Valuation → Stock sources for valuation. Inspect positions, recorded quantities, accounting units and the portfolio’s exact economic cutoff. The report belongs to that ledger and cutoff, not automatically to today’s price.

The new stock-source assignment is separate from existing source mappings. It changes no recorded trade, quantity, cash or cost. Every position remains in the report even when it lacks an admissible price.

2. Select an exact candidate and review the input

Use Select owned position to choose an actual position, then Load catalogue and choose exact CIK, ticker and original exchange including unknown. Missing or ambiguous matches must not be replaced by a ticker alone or a similar name. A source row ordinal is not durable identity.

Load listing and history prepares the selected native input and, alongside the SEC candidate, review returned symbol, name, venue, instrument type, currency and capture. A matching name or ticker alone does not prove a price-to-CIK issuer join. Assigning it to an owned position is your explicit assertion.

3. Confirm the unit and separately save the assignment

Confirm that this position’s unit corresponds to one share of this listing, then choose Apply assignment; a CFD, multiplier or other unit is not inferred. Preserve exact source currency case. GBp is neither GBP nor USD; the application applies no FX or 100:1 scale conversion.

The private assignment stores only the chosen candidate descriptor and unit. Review the exact position before confirming the save. Prices, histories, source documents and financial report are not the saved configuration; reading or saving an assignment does not itself calculate a report.

4. Explicitly calculate the report and read coverage

After reviewing inputs, choose Calculate stock valuation at the portfolio cutoff. A daily timestamp alone does not establish availability of its Close: a point in the currently reported regular session needs that session’s end at or before cutoff; other points conservatively need 24 hours after their native timestamp. This is an application rule, not a verified historical market-close time. Selection uses the latest such eligible completed point, at most seven days old.

If the latest eligible point has null, an older non-null Close is not substituted. An over-age point, missing source, unclear unit or a later recorded ledger split or reversal changing the basis is a specific coverage gap. An unavailable value is not zero and no position silently disappears.

5. Separate values, cash and cost PnL

An admissible position value uses exact quantity × original Close, without binary financial arithmetic. Subtotals use exact own units. Cash is shown separately; mixed currencies are not combined into an invented base total and nothing is automatically converted. A base total is admissible only with complete coverage and exact agreement of every price/cash unit with the base currency.

Cost PnL is available only with an exactly compatible accounting unit and supported basis. It is not total return, TWR, XIRR or a tax calculation. Missing or incompatible costs receive an explicit unavailable reason instead of an estimated gain.

6. Inspect the original point and export the whole report

Summary values, whole table and point detail belong to one report with the exact owned ledger revision/hash, assignments and originals. Inspect the selected timestamp/Close, source URL/SHA, unit, capture and coverage reason. A ledger, account or source change requires another review.

Whole JSON and readable CSV retain every position, cash balance, subtotal, gap and original source. Pages and detail do not narrow the file. Before download, current owner, session, ledger basis and all same source originals are checked again; an invalid constituent blocks export rather than substituting a newer price. The file includes your private ledger basis, not just market prices.

The file is neither a canonical financial Snapshot nor public redistribution authority. This valuation adds no historical performance, FX, tax results or proof of listing identity.

1. Retain a particular original result

First explicitly load and apply a single listing history or a comparison of 2–6 candidates in the catalogue. Review every actual identity, currency, range, window and clock. Saving the original result binds the whole applied original, rather than draft edits or the currently visible table page.

Separately prepare the original result for saving, name it, review it and confirm an immutable Snapshot. Original source documents and every selected or excluded point are retained; saving fetches no newer values. If the exact temporary input can no longer be verified, saving is refused and another source never silently replaces it.

2. Open the same Snapshot from My analyses

The original entry opens the earlier graph, table, detail and whole JSON/CSV after current account and access checks. Opening or exporting requests neither a new Yahoo history nor a new SEC catalogue. The table may be paginated; the whole export retains the original result and inputs. When verification expires, separately verify that same saved original.

To work with current data again, adopt conditions only into a new draft and explicitly confirm replacing unfinished work. The transfer carries no financial inputs and itself loads or calculates nothing. A subsequently loaded history is a new result; the old Snapshot is not overwritten.

A SEC candidate and a native Yahoo listing are distinct evidenced identities; a matching ticker does not prove the share class or issuer join. Original currency and clocks remain intact. AdjClose is not an automatic total return, and a private Snapshot grants no public distribution rights.

3. First apply an actual detail or comparison

In the stock catalogue, load the selected input, review native identity, currency and source history range, then apply the detail or comparison of 2–6 listings. Apply any local date window separately. Saving uses the last applied selection, not subsequently edited drafts.

The configuration retains exact CIK, ticker and original exchange including unknown, chosen range, local window and page size. Original row ordinal, native Yahoo identity and currency are not stored as evidence. Review them again in the newly loaded input after returning.

4. Name and confirm the private save

Save applied settings uses the last applied analysis. Sign in, enter Analysis name and choose Review save. Inspect the detail or complete ordered comparison, range and window. Only Confirm save to account submits your private configuration. Saving itself neither downloads data again nor calculates anything.

A settings entry copies no prices, OHLCV, original documents or computed series. Use the separate original Snapshot save for a particular history or comparison. A private save is not public sharing.

5. Find, rename or remove your own entry

My analyses and Return to work expose your own named stock configurations. Select the current entry; rename and remove require their own review and confirmation. Removing a configuration deletes neither the account, provider originals nor other analyses.

Entries belong to the current owner and session. An older command receipt can describe its original write while the library has a newer state. After an uncertain response, retry only the exact retained request; the app must not automatically create a second save.

6. Open current settings and adopt a draft

Opening a saved entry verifies the current account and current settings revision. Review it and choose Adopt as draft. An unfinished workspace must not be silently replaced: confirm Replace unfinished draft, or keep your current work. A passively changed or removed entry requires another review.

Neither opening nor adoption loads the catalogue, histories or prices, or runs a calculation. Renewing a short proof for the same revision preserves valid review; expired proof needs renewal. A draft is not an admitted financial result.

7. Explicitly load current inputs and resolve matches

After adoption, choose Load current catalogue first and resolve the exact saved candidate descriptors. A missing, ambiguous or changed candidate is not replaced by a similar name or a ticker alone. Select and review an actual input again.

Load listing and history or Read selected listings prepares the new originals. Review every native identity, currency and actually returned range, then apply them with one confirmation. The configuration retains no earlier currency; different comparison currencies are not converted automatically. Saving is neither current authority nor a new price.

8. Keep the new result separate from the configuration

The chart, full table, point inspector and whole JSON/CSV then belong to the newly applied result and its source clocks. A date-window selection is local; original points outside it remain retained. Returning to a name creates no historical data Snapshot and supplies no earlier source values.

Account changes, hiding or leaving the workspace block old private actions and late replies. Saving configuration itself grants no new data rights, public approval or listing-identity verification.

1. Choose on this device

In Settings, open App on this device. First inspect what is stored locally: only a fixed static shell without account or data. Enabling is your explicit choice for this browser and device; it publishes nothing, creates no account and requests no notification permission.

If you do not want the feature or your browser does not support it, continue using Datisto on the web. Installation is optional and grants no paid or data permission.

2. Enable the shell and use the actual browser option

Choose Enable app in this browser and inspect the reported state; later use Check status to verify it again. If the browser offers an installation dialog, Install app opens it for your decision. Dismissing it is not a successful installation. The app does not promise the same installation API in every browser.

Without a supported dialog, use the available manual browser-menu instructions, such as installing the site or adding it to the home screen. Showing instructions does not confirm installation; if the option is absent, keep using the web. Account and sign-in remain the same.

3. What being offline means

The offline shell only states that the live app is unavailable. It contains no account, financial values, portfolio, community words or private drafts. This shell does not retain APIs, sign-in or fetched financial data for offline use. It supplies no offline authority, old prices as current values or queue of unsent writes.

If connectivity is lost during open work, the financial and private workspace is concealed. Unsaved in-memory work is not silently deleted, but it is not backed up by this feature. Do not submit changes based on an old connection or old authorization.

4. Return to live work explicitly

Connection is back does not itself reveal the previous account or confirm current source data. Prepare fresh reload opens a warning about losing unsaved in-memory work. Choose Keep unfinished page, or explicitly Discard unsaved changes and reload. Only a fresh load checks the current session and rights again.

There is no automatic synchronization or replay of earlier writes. If the outage continues, the truthful offline state remains; the installed shell is not a substitute financial dashboard.

5. Disable the local shell and distinguish installation

Disable app component removes only Datisto’s own registration and exact static cache, not unrelated browser components or all storage. It deletes neither the account, server data nor a system-installed icon. An already-open page can remain controlled by the original component until another load or closure.

Removing the icon or installed app itself is handled by the browser or device system. A standalone window, enabled shell and confirmed installation are different states. Native installation across devices, mobile quality and public operation require their own verification.

1. Build an exact roster

The US stock catalogue loads automatically; use Pin for comparison on any additional rows you want. The roster identifies exact rows and their catalogue capture, not just a name, ticker or CIK. One CIK can have several share classes or listings; none is automatically merged.

Pin two to six rows. Reorder them with the arrow controls, remove individual choices or clear the roster. Search and filtering do not load prices for the whole catalogue. Changing the pinned selection reads only its at-most-six members. A catalogue selection does not prove that every listing is available.

2. Choose a history range

Choose 1mo, 3mo, 6mo, 1y or 5y with daily interval 1d. The selected roster loads automatically, with at most two reads at once. Review every input’s native symbol, name, venue, instrument type, currency, actually returned range and capture. Availability of all five ranges for every listing is not assumed.

The SEC candidate and native Yahoo listing remain separate source records. Matching a name or ticker alone does not prove CIK issuer or share-class identity. Failure of one member neither shrinks the roster nor substitutes a similar symbol; the previous dated comparison remains retained.

3. Read the whole validated comparison

The whole comparison appears after every native identity, currency and range is checked. A failed member is never silently omitted. Changing the selection or range cancels the old request and reads the new one; Refresh comparison explicitly repeats the read.

Every member must have exactly the same native currency token, including letter case. USD, GBP, GBp and GBX are neither interchangeable nor converted. A dimensionless base-100 curve does not permit mixing currencies.

4. Select a local time window

Within the captured input, enter From and Through inclusive; both dates are included by the UTC date of the original timestamp. Apply time window creates the new applied view. Whole captured period removes the local restriction. Window changes download nothing and do not change the history range requested from the provider.

A window may contain gaps or too few common points; missing days are not invented and prior prices are not substituted. A new comparison basis is sought inside the window. Every original point, including those outside the window, remains in the whole table, evidence and export with its exclusion reason.

5. Read common points and gaps

The base of 100 is the first shared completed exact source timestamp inside the applied time window at which every member has a positive Close. At least two common usable points in the window are required. Each curve uses original Close divided by that member’s Close at the shared basis; the basis is not moved to a different day for each stock.

The calculation 100 × Close / own basis uses an exact fraction and one HALF_EVEN rounding to at most 24 decimal places. The basis is 100; a genuine zero Close after the basis remains zero rather than a gap.

Exact native timestamps are joined, not nearby trades or merely matching UTC dates. An absent timestamp, null Close, current session and pre-basis point remain distinct; none is replaced by zero or a previous price. The table offers all, common and excluded points, pagination and original point evidence.

This compares the original Close series, not total return with dividends, fees or taxes. Separate Adj Close does not replace the input. A daily timestamp is not a confirmed closing trade time; a current quote has its own timestamp. This view derives no volume sum, FX, correlation, alpha or portfolio.

6. Inspect evidence and export the whole comparison

The chart, complete table and point detail share one applied query, result and manifest. In evidence, inspect every native identity, original Close and basis, exact timestamps, SHA-256 and individual source captures. Cache describes application reuse, not current market validity.

Whole JSON and readable CSV retain every original input, basis, common and excluded point, including all points outside the time window, and its reasons. Neither the current page nor selected detail narrows the file. Before download, every same retained original is checked through the current local gate; a missing or expired member blocks the whole export rather than substituting a newer price. A new set requires an explicit read.

This local analytical file is not a named account configuration, canonical Snapshot, ValuePassport or public share. Catalogue coverage is not a count of histories already acquired.

1. Load the whole catalogue and apply a selection

The catalogue loads on its first visible opening; its capture date appears beside the table. Search ticker, name or CIK; filter the exact original exchange label, including unknown exchange. Sort by original order, name, ticker or exchange and choose a direction. Apply selection confirms the conditions; pages of 25, 50 or 100 rows do not reduce the source coverage.

This is the whole particular SEC list, not every stock worldwide or a guarantee of current trading. Row counts can change between captures. Multiple tickers with the same CIK remain separate, including different share classes. Loading, searching and pagination do not request prices for thousands of securities.

2. Open an exact listing

Selecting a ticker automatically reads its bounded Yahoo daily history and internally validates the exact catalogue row and source identity. The ticker is not rewritten, punctuation is not guessed, and a similar company name is not substituted. Only an exact row from the admitted catalogue may select the source.

To change the range, choose Source history range: 1mo, 3mo, 6mo, 1y or 5y, with daily interval 1d. The response supplies the quote and original history for that source range. Inspect the actual returned start and point count; five years are not guaranteed for every listing. Changing the source range reads that range; the result always identifies the period actually returned. An unavailable candidate is not priced at zero and is not replaced by another symbol.

3. Review native identity and currency

Alongside the SEC candidate, review Yahoo’s returned symbol, own name, venue, instrument type and currency. The SEC candidate and native listing are separate source records: matching a ticker or name alone does not prove that the price belongs to that exact CIK issuer. Do not automatically merge share classes or venue listings.

After internal identity, currency and range checks, the public result appears directly without a second quote-approval step. Currency and letter case remain exactly as reported: USD, GBP, GBp or GBX are not automatically relabelled or converted. No default currency is substituted for the price unit. Missing identity or units prevent an unverified price being presented as admitted.

4. Distinguish the quote, Close and Adj Close

The regularMarketPrice quote has its own timestamp. It is not automatically live or executable, and that timestamp is not the response download time. Previous Close and chartPreviousClose remain separately labelled source fields; the application does not invent a daily change from their names. Also inspect the venue, timezone and reported regular session.

Daily Open, High, Low and Close are original provider values; Adj Close is a separately reported series, not an automatic replacement for Close. These fields do not establish our own corporate-action adjustment, reinvested dividends or total return after taxes and fees. Volume is its own reported quantity, not a count of people or dollar trading value.

5. Apply a local time window

Within the captured input, enter From and Through inclusive; both dates are included by the UTC date of the original timestamp. Apply time window creates the new applied view. Whole captured period removes the local restriction. Window changes download nothing and do not change the history range requested from the provider.

The table distinguishes All original points, In window and excluded points. Missing trading days are not filled. The local filter changes neither the original quote nor its own timestamp; whole JSON/CSV still retain every captured point, including those outside the window.

6. Inspect the chart, whole history and a point

Price and volume charts, the complete paginated table and point detail use the same applied input. Detail preserves original order, exact OHLCV and separate Adj Close. Zero, null and missing information are not interchangeable; source gaps are not filled and weekends are not invented as new points.

Inspect whether a row belongs to a completed or current session according to the reported start/end and capture. A daily timestamp is the original source time of a daily point, not a confirmed closing trade time; a current point can fall inside the regular session. Original days and counts do not certify a complete expected trading calendar.

7. Check source evidence and export the whole original

Evidence separates the SEC catalogue from the selected Yahoo chart. Inspect URL, SHA-256, request start, completed capture and available HTTP Date. Cache describes how long the application may reuse a response, not how current the market price is. A failed read preserves the previous dated view; another load is explicit.

1. Select an actual pair in the original list

In Binance Spot, select USDT or BTC quotes, apply search and open a row detail. Review the exact native symbol, base asset and quote asset. BNBUSDT and BNBBTC are different markets; symbols are neither translated nor substituted with similar names. Original symbol characters are preserved.

Opening an actual pair automatically reads its bounded daily history. Pairs must belong to the admitted original list; browsing and searching that list do not download every pair’s history. If the list can no longer be verified, refresh it and select the pair again.

2. Read the loaded history and its input

On first opening, up to 90 days load and history appears after internal checks of the exact pair, unit, original candles and actual range. The result retains its capture time. A read may reuse an original retained response; it does not establish a new quote right now. The 7, 30 and 90-day presets directly change the local chart and table without another source request; Apply custom endpoint dates. Accept a refreshed input explicitly when already editing your own selection.

A source failure, rate limit or cancelled read does not replace the preceding dated result with demo data or another provider. Inspect its stated status; another attempt is explicit. Automatic retries and fetching all pair histories are not enabled.

3. Select completed UTC days

Choose 7, 30 or 90 days, or a custom window within the captured latest 90-day range. Both entered From inclusive and Through inclusive dates are included. The exact query uses the following UTC day as its excluded end. Explicitly apply a changed window; it changes the view of the same input without downloading again. The available end is anchored to the original capture, not today’s clock when reopened.

A day still open at capture is not a completed daily value. Its original row remains in excluded evidence. A missing day or a newer pair’s shorter history is not zero and is never filled with the last price. Inspect actual gaps and available-day counts.

4. Read price and volume in their own units

Open, High, Low and Close are native quote prices per one base unit: for example, USDT per 1 BNB or BTC per 1 TRX. USDT is not relabelled USD, and a BTC price is not a quotient of two dollar prices. Base-asset volume and quote volume are separate quantities; do not sum volumes across different pairs or quotes as one unit.

Daily history is separate from the rolling 24-hour statistics in the main table. Candle open/close times are not the last trade time, publication time or a currently executable quote. Trade count is not a count of people, and taker-buy volume does not identify buyers.

5. Inspect every original row

The chart and complete paginated table use the same applied window. Open a candle for all twelve original fields: times, OHLC, base volume, quote volume, trade count, taker-buy volumes and the uninterpreted final field. Detail preserves exact numeric strings and original order; graph display does not replace their precision.

Evidence distinguishes the pair’s original market identity from its separate history acquisition. Inspect the source URL, SHA-256, request start, completed capture and available HTTP Date. Application reuse validity is neither venue quote validity nor proof of a complete historical market.

6. Download the whole same result

JSON and readable CSV retain the applied query and manifest, exact pair identity, every original candle and matches/exclusions. Table page, selected day or a narrow window do not shorten the file. Before download, the same retained original and current local source admission are rechecked; export fetches no new history and substitutes no newer price.

Apply the catalogue first, then save settings

In the fund catalogue, load and apply an input, edit filters or a selection of up to six products, then confirm Apply. Save applied settings saves the last applied query, rather than unconfirmed filter edits. It also retains 25, 50 or 100 rows per page and the exact inspected product ID; this product can be outside the compared selection.

After signing in, enter a name, choose Review save and inspect the exact conditions, selection and presentation. Only the separate Confirm save to account sends the change. Financial data, prices, source catalogue and result are not stored. This is private configuration for the actual catalogue, not a financial Snapshot.

Find the selection in My analyses and return to work

Saved fund selections appear in My analyses and in compact Return to work entries. The library can search names or products, prepare rename or removal, and separately review and confirm each change. Names and collection are available only after current verification of your own account and session.

The link opens the exact saved-settings identity. Inspect its original conditions, then choose Adopt reviewed conditions into draft. Opening and adoption fetch no issuer data and calculate no result. Active local work is not overwritten; adoption requires explicitly discarding your draft. Review a changed saved version again; a removed entry cannot be adopted, and expired verification requires account re-verification first. Renewing verification of the same version preserves a valid review.

A new calculation, missing products and uncertain saves

After adoption, load a catalogue and explicitly Apply it. Source age and availability are reevaluated; old settings imply neither retained prices nor present availability. Missing selected IDs remain in the draft and prevent silent application. The app neither substitutes other products nor drops them automatically. An unavailable inspected product is identified by its exact ID.

Short private verification renews during active visible work without erasing save review. Account changes, hidden pages and late responses cannot confirm another actor’s request. An uncertain outcome retains the exact key and contents; retry requires an explicit control. Leaving the page does not cancel an accepted write. The catalogue’s complete result export is separate and includes source evidence; account settings do not.

Your own CSV, with every column explicitly assigned

In Portfolio → Imports choose My CSV columns. Upload one file or paste comma-delimited text. This accepts bounded inert UTF-8 text, at most 500,000 bytes and 2,000 data records. A quoted multiline cell is one logical record; XLSX, ZIP, macros and formula execution are not used.

Original positions identify columns, so two equal headings remain distinguishable. Explicitly choose one universal field or Ignore for every column. No output field can be mapped twice. Economic time and the original supported event type are required. The account comes either from an explicitly mapped account_id column or from your chosen own account for the whole file.

Literal conversion without inferred economics

Mapping changes column arrangement and names. It guesses no date, time zone, decimal separator, sign, currency, event type or instrument. Date and numeric strings are copied literally; the existing universal kernel checks validity. A locale date, decimal comma or custom word for a purchase is not silently rewritten. Empty values, explicit zero and the text null are distinct.

Before preparing a server preview, compare original records with generated universal rows. Inspect explicitly ignored columns too. Changing the file, mapping or account invalidates the previous review. Two legitimate identical operations are not merged by conversion. A stable operation_id helps establish duplication; equal amounts and dates alone do not.

Retained lineage, preview and atomic confirmation

Server staging binds the original file, explicit mapping and generated universal CSV hash. Preview shows new, duplicate, conflicting, ignored and erroneous records and warnings. Only your separate confirmation of the whole acceptable group writes the ledger. Confirmation rechecks original lineage and conversion; upload or review alone posts nothing.

An uncertain outcome retains the original operation’s exact identity and content. Verification and retry are explicit; no different new request replaces the old one, and cancelling a wait cannot roll back an accepted write. Ordinary renewal of the portfolio’s short verification preserves drafted mapping. Account or portfolio changes and late responses are isolated.

Confirmed imports connect to existing history and supported import Undo; Undo uses opposite entries and has its own conditions. Mapping your columns establishes no history completeness, cost basis, market price or correct tax output. XTB, IBKR, Trading 212, DEGIRO, Fio and eToro still require genuine permitted anonymized samples and verification of each format. This is no automatic broker detection, AI mapping or broker connector.

Open your own report and read the information request

In My data correction reports, choose your submission. A fresh account check admits the original explanation and link, current status, report revision and recorded review steps. If status is Needs information, read the actual request in history. Received means a recorded submission, not correction approval or a processing promise.

Supplements are available only for Received, Needs information and Under review. Rejected or duplicate reports accept no new evidence. Up to 50 supplements remain separate immutable records with their own dates and revisions.

Write, review and explicitly confirm

Write 20–8,000 characters of additional explanation and a primary-evidence HTTPS link. Choose Review supplement. Inspect the exact text, link, target private report and expected revision. Confirm and append evidence sends the write. Back to draft sends nothing; ordinary detail-proof renewal does not erase drafted text.

The original explanation and link are never overwritten. Appending evidence advances the report revision but does not itself change review status, publish a community post, correct a price or endorse the evidence. Acceptance is confirmed only after a verified receipt and fresh own detail contain the same accepted supplement.

Revision changes, expired verification and uncertain outcomes

Private detail has short-lived verification. Expiry conceals content until renewed verification; ordinary renewal of the same report preserves its draft. If the revision changed during review, verify it and review the supplement again. Account, report, route or visibility changes invalidate held controls and late responses.

An outage can leave acceptance uncertain. The exact retained request in browser memory includes original text, link, revision and identity. Verify by retrying the same supplement sends only that request and can verify its original acceptance without a second record. There is no automatic replay or new request substituted for an uncertain one. Cancel waiting cannot roll back an accepted write. Closing the whole browser can lose temporary memory; verify your own history first afterward.

Daily history for the selected pair

Open an actual pair detail; its bounded daily history loads and appears automatically. Supported pairs must belong to the original admitted list; a list does not mean every pair history has already been acquired. Completed UTC daily candles are separate from rolling 24h statistics and retain USDT or BTC quote units. The pair-history guide explains later period selection.

Original venue coverage, rather than the whole crypto market

The overview uses Binance spot directly: the captured product catalogue with TRADING status and spot eligibility, and original FULL 24h statistics. Eligible coverage contains USDT- or BTC-quoted pairs joined by exact venue symbol. Pair counts are not global token counts, capitalisation or a whole-market ranking. A base-asset code alone does not identify an equivalent token in another catalogue.

This independent source view neither replaces nor bypasses CoinGecko aggregate prices or admission gates. The first visible opening reads original inputs once and applies the default USDT overview. No demo prices, assumed token logos or successful human activity are fabricated.

One quote asset in each applied query

Choose USDT or BTC, a symbol or base-code search, sort field and direction. Confirm with Apply conditions. Until confirmation, table, detail and export retain the preceding applied query. Page sizes of 25, 50 or 100 and page navigation affect presentation only. Search preserves native characters, including non-Latin venue codes.

USDT is not USD. Prices are quote assets per base unit; base volume and quote volume have different units. Price and quote-volume ordering stays within the one selected quote asset. Volumes are neither summed across unlike pairs or assets nor converted to dollars, and no market capitalisation is produced.

Read all 21 original fields

Select a row for original product identity, last returned price, source percentage and absolute change, weighted average, Open/High/Low, bid/ask and quantities, base and quote volumes, trade counts and original IDs. A missing ticker remains empty; an explicitly supplied zero stays zero. Exact decimal strings remain in detail; ≈ marks presentation rounding only.

openTime and closeTime delimit the provider’s rolling statistics window. The window end is not last-trade, publication, revision or quote-validity time. Bid/ask is no atomic execution guarantee. No intraday or seven-day chart is fabricated without a historical series. Separate original daily history is linked only for the already supported BNB/BTC and TRX/BTC pairs.

Coverage, refresh and a whole verified export

Cards distinguish the whole admitted pair-identity cohort, pairs with the chosen quote, applied-condition matches and available or missing statistics among those matches. Excluded identities and source positions remain retained. Manual Refresh prepares a new set for the next Apply. Failure preserves the preceding dated overview; automatic polling is off.

Opening evidence shows separate read start, completed capture, HTTP Date and SHA-256 for both sources. A whole multi-megabyte raw body is displayed only when explicitly opened. Five-minute reuse starts at the earlier capture and one-hour retention at the later capture; these describe application reuse, not current market validity.

Venue pair and its unit

BNB/BTC means BTC per 1 BNB and TRX/BTC means BTC per 1 TRX on Binance spot. These are not ratios of two aggregate USD prices, whole-crypto-market indices or new CoinGecko catalogue identities. Original Binance product metadata in the detail preserve the symbol, base and quote assets, and product status at capture.

Each pair has its own chart and scale. Daily Close is the original close of a specific candle. It is not a current executable quote, guaranteed execution price or simultaneous latest trade across the pairs. Product and two history captures keep separate clocks.

Choose the period and pairs, then Apply

The first visible opening reads the original product document and two bounded daily histories once. It applies 90 preceding UTC days according to the captured inputs when admitted. The 7, 30 and 90 day controls, custom dates and selection of one or both pairs change only a draft; confirm with Apply. The period is start-inclusive and end-exclusive, up to 90 days.

The calendar retains every selected day, including absent or unusable candles. Lines do not bridge gaps, smooth them or carry a preceding price forward. The current source day’s open candle is not treated as a completed daily close. Unknown observations are not zero.

Inspect the whole original candle

Select a chart point or table day. Detail includes original Open, High, Low, Close, base-asset volume, BTC quote volume, trade count, taker-buy volumes and all 12 native candle fields. Candle opening and ending labels remain original venue time fields; they are not last-trade, publication or revision timestamps.

Collapsible evidence preserves all three original responses, SHA-256, read start, completed capture and HTTP Date. Excluded open candles, out-of-window observations and the unused pair remain in the whole model and export. A public source link does not mean its current response will match the held captured body.

Refresh and export explicitly

Automatic polling is off. Manual Refresh prepares new inputs for a subsequent Apply. Failure preserves the preceding dated result. Cache time describes application reuse of the response, not market currentness or quote validity.

Two distinct measurements and separate charts

Confirmed transactions are transactions the provider reports as included in blocks. They are neither pending mempool transactions, economic payments, settled value nor people. Unique addresses used are addresses under the source’s definition; a person can use multiple addresses and an address can serve multiple users. These are not counts of wallets, investors or users.

Each chart retains its own unit and scale. Transaction and address counts are not combined, and summing daily addresses would not give period-unique addresses. Lines connect only adjacent available days; gaps have no interpolation, carried-forward value or smoothing. Missing observations are not zero.

The period ends at capture, not the last value

The first visible opening reads two original sources once and applies 90 preceding source days up to the capture date when evidence exists. The provider may return its last value several days earlier; the missing tail remains in the calendar. Cards show the latest available day in the applied period and subsequent days without a usable observation.

The 7, 30 and 90 day buttons only prepare a draft. Custom intervals are start-inclusive, end-exclusive and at most 90 days. Choose one or both metrics and confirm Apply period and metrics. Until then, charts, table and export retain the preceding query. Changing the table page or selecting a point does not change the result.

Complete calendar and exact original point

The calendar retains every selected day, including days wholly absent from the source. It distinguishes an available count, original null, absent day, unavailable source and an unusable negative, non-integer or oversized count. A valid zero remains zero. Select a count or gap in the table; available chart points can also be selected with the keyboard.

Details preserve the original UNIX x, exact y token, source position, native metadata, read start, completed capture, HTTP Date, SHA-256 and whole original response body. For example, 736499.0 can qualify losslessly as integer count 736499; its original spelling remains alongside it. Point dates are source UTC labels. Exact aggregation boundaries, publication, revisions and finality are not independently established.

Refresh and export the same whole result

Automatic polling is off. Refresh original inputs prepares a new set for explicit Apply. Failure or cancellation preserves the preceding result with its original clocks. The six-hour cache starts at each source’s completed capture; it does not prove network freshness or a final daily count.

Companies and classification, rather than sector ETFs

The company industry map contains a fixed dated selection of 30 individual companies. Groups use the SIC code and original description in the dated original SEC document. This is neither GICS, the entire equity market nor a historical classification reconstruction. Classification publication and effective dates are not supplied; a price-session date does not establish them. Dated catalogue identity and declared filing identity remain separate; mismatches are not rewritten.

Every company has an equal-size tile. Area does not represent market capitalization, company weights or an aggregated industry return. The separate Sector ETF map remains a view of eleven fund proxies; the two maps are not interchangeable.

One completed session and adjacent original closes

The first visible opening reads the existing price batch and original classifications once. If the inputs permit a common completed session, the latest choice is applied once. Further session, industry, direction and search changes are drafts confirmed with Apply. Until then, map and table retain the previous applied query. Sorting and switching Map/Table affect presentation only.

Change = 100 × (selected original Close − immediately preceding original Close) / preceding Close. Both points must be completed; a missing price is not skipped to an earlier usable value. A shared timestamp identifies the native source-session start, not a simultaneous closing execution. Original Yahoo Close includes split adjustments; it is neither Adj Close nor dividend total return.

Missing inputs, company detail and full export

Missing prices and missing classifications are different states. A company without classification remains in a separate group. Original price points can be inspected separately; without a verified identity association the derived change remains unavailable. An unavailable change is not zero. Choose a tile or row for original SIC, dated SEC evidence, the two specific Yahoo prices, their positions, capture clocks and SHA-256. Technical evidence is collapsible; profile and price links open separate analyses.

Manual refresh reads inputs. Failure retains the clearly dated prior result; success prepares a new set for explicit Apply. JSON/CSV contain the same whole model of all 30 companies, matches and exclusions, original inputs and manifest. Export verifies current local access without reading new prices or SEC documents. This is not a canonical account Snapshot, investment advice or a measure of the whole market.

Initial period and your selection

Opening automatically prepares and reads at most 90 calendar days from stored BTC and CNB coverage. Both endpoint days are included in UTC. Edit and Apply a custom interval; no different source or rate is substituted.

A converted point is the original Coinbase BTC/USD daily close multiplied by original CNB CZK per USD with the same source-day date. This is a source-converted close, not a direct venue BTC/CZK quote, simultaneous exchange offer or return. Open the separate Coinmate tab for an original direct quote; its sources and clocks differ from the daily conversion.

Chart, gaps and both original Passports

The chart shows only converted CZK points. The table retains every day in the period, both original inputs and any unavailability reason. Missing-fixing days, including weekends, remain gaps: Friday’s rate is not carried forward, interpolated or replaced by zero. Original candle open/high/low and volume are not converted.

Choose a day for the exact decimal result, separate clocks and both original Passports. A BTC daily candle completes at a different time from CNB fixing; the same date does not imply simultaneous prices. Capture, ingestion, publication and possibly unknown freshness remain separate. Calculation rounds to at most 18 places using HALF_EVEN; display labels may be shorter.

The direct Coinmate quote is a separate source snapshot

Direct quote reads the original responses once on its first visible opening. Further Refresh actions are manual; returning to a visible tab does not refresh a retained snapshot or retry a failure. Last, bid and ask are original Coinmate BTC/CZK values in CZK per BTC. One snapshot is not a historical series; no past-price line is invented. Other original response fields remain in the inspector without inferred windows, units or formulas.

The endpoint time is source-supplied, not a verified last-trade time. Completed capture of both responses, request start and one-minute application-cache expiry remain separate. Cache expiry does not establish venue quote validity. Publication, last-trade time and quote validity remain unknown.

Refresh is explicit; no automatic polling runs. Failure retains the prior snapshot with its original time and a refresh error. JSON/CSV first verify access to that same retained original, without calling Coinmate again or replacing its quote. Eviction or restart can make export unverifiable; reading a new source requires your explicit action. This source snapshot is not a canonical account Snapshot, saved history or value Passport.

The exchange-books link opens the existing BTC/CZK depth workspace with your amount and fee assumptions. A ticker does not promise execution of an entire trade at the displayed price.

Choose the source, unit and method

In Followed metrics, search the current catalogue by measurement, country, unit, frequency or source. Follow stores an exact series reference, rather than an entire entity, a current price or copied data. Local lists hold up to 30 entries, with your ordering and explicit removal.

A selected measurement opens its original chart, table, value Passport and export over the same result. The existing analysis controls the period and knowledge time. Different units are not combined. This page does not bypass special source-cohort admission.

Local and account selections are separate

Without signing in, only configuration is stored in this browser. Signing in neither erases it nor sends it automatically to an account. Under My account and merge, first open the review. Account order comes first, exact duplicates are not repeated, and differing identities for the same series require choosing the account reference, local reference or removal from the merged result.

Nothing is truncated above the limit of 30; edit the selection first. After reviewing, explicitly confirm the change. Local selections remain intact after success. Account membership and ordering also change only after review and confirmation.

Verification, recovery and unavailable measurements

Private selections are visible only while the owner and session are currently verified. They are concealed after switching accounts or verification expiry. If a save outcome is unconfirmed, explicitly verify and retry the retained request using the same key. Reloading does not send it automatically; recovery depends on this tab’s storage.

Missing, changed, inaccessible or unsupported sources remain marked in the selection. The app never substitutes an entity’s first series. Resolve a changed identity by explicitly removing the old reference and following the new one. Following grants no data permissions. Original clocks, units and unknown freshness remain visible in the analysis.

Default issuers and your selection

On first opening, a bounded queue reads the latest actual filings for the three default issuers. Apply confirms your own selection and conditions. Catalogue names and symbols do not verify current listings.

Conditions belong to the whole applied query

Choose tables, transactions or holdings, and optional original codes. P/S may be market or private transactions; A, F, M and G have distinct meanings. An empty code selection also includes rows without a code. Dates can refer to the transaction or filing; interval start is inclusive and end exclusive.

A missing transaction date remains unknown. Including undated rows is explicit; report or filing dates never replace it. Excluding 4/A amendments marks those original latest rows as excluded; an older Form 4 is not substituted.

Inspect the whole coverage roster

Coverage retains every selected issuer, including unavailable sources. Latest filings can have different dates and captures. An issuer with no supplied form does not have zero activity; a filing declaring another issuer is not assigned to the selected company. No matching rows does not mean no actual transactions.

Inspect matches, exclusions and original context

The ledger shows matched, excluded or all source rows, with exclusion reasons. Search, sorting and pagination affect presentation only. Detail retains the exact issuer, accession, row type and source ordinal, every original field, container and field footnotes, and remarks.

Reporting owners belong to the whole filing. A joint form does not assign every row to an individual, and rows are not multiplied by owner count. Open this exact original filing uses its CIK and accession, rather than a newer substitute form.

The scope is latest originals, rather than complete history

The screener calculates no quantity-times-price, net flow or aggregate purchases/sales. It does not establish complete recent activity, actual publication time, current holdings or economic ownership.

Choose an original filing

Insider filings automatically reads the default issuer’s actual filing. Find another supported identity by name, symbol or CIK and open its latest or exact filing. Search does not read all 30 issuers. The symbol/name/CIK association is a dated catalogue rather than a current-listing claim; the original filing name and symbol remain separate. A 4/A amendment stays separate with its original filing date; the application does not automatically merge it with an older report.

Read the event by its original code

Read the code together with the acquisition/disposition direction. P/S cover market and private purchases/sales; awards, option exercises, tax settlement and gifts have their own codes. Original prices, including zero, retain their explanatory footnotes.

Inspect both security tables

The combined ledger retains transactions and separately reported holdings for non-derivative and derivative securities. A holding without a transaction date remains undated. A supplied empty table and an absent table stay distinct.

Open every footnote and reporting owner

A row detail shows original values, direction and ownership form, including footnotes attached to individual fields and row sections. Reporting owners and their issuer relationships belong to the whole filing; a joint filing alone does not assign every row to a specific owner.

Inspect clocks and download the whole

Event date, report period, original-submission date, SEC acceptance and document capture are separate clocks. Search, sorting and pagination change presentation; whole JSON/CSV retains every original row, footnote, source and applied result manifest.

The example uses fictional identities

An unavailable actual filing stays unavailable. No fictional owner, transaction or zero report replaces it.

Choose a manager and filing

Opening loads the default manager’s actual filing. You can then choose another manager and its exact quarterly filing. The latest filing may be a 13F-NT notice without its own table; an older report is not substituted.

Distinguish the original clocks

Period end, filing date, SEC acceptance and document capture are separate clocks. Reported holdings are neither today’s portfolio nor a record of exact trades.

Inspect every reported row

Search, sorting and pagination change presentation only. Repeated CUSIPs stay separate; class, SH/PRN, options, other managers and voting remain on the original row. A CUSIP does not establish a verified ticker.

Read value and scope correctly

The supported current format reports values in USD rounded to dollars. Matching cover counts and totals checks this table, rather than establishing all manager assets. Amendment scope and any confidential omission stay visible.

Inspect sources and download the whole

A row’s dossier retains original values and filing context. Inspect source documents, hashes and age. JSON/CSV contain the whole applied result, independently of search or the open row.

Compare two consecutive quarters

In quarter comparison, select two exact filings from the same manager and apply the selection. A group combines the exact same CUSIP, class description, SH/PRN and option; its detail retains every original contributing row from both reports. A changed class description is not automatically treated as the same identity.

Newly and no-longer reported groups describe presence in the table. An absent side is not zero ownership. A change in reported quantity does not establish a purchase or sale, and a USD value change is not price return. Notices, amendments, confidential omissions or incompatible scope have an explicit calculation-unavailability reason; original inputs remain in the whole export.

Select groups across multiple managers

In the combined view, choose managers, exact filings for the same quarter and an operation: union, intersection or a minimum reporting-manager count. Each manager counts at most once per group; all its original rows remain in detail. Apply the selection explicitly.

A notice or unavailable table stays unknown coverage in the denominator. Results show lower and upper possible counts and groups with undetermined membership. An incomplete observed list does not establish an empty intersection. Reported amounts and weights belong to individual reports; the combined selection is neither total assets under management nor a trade recommendation.

Original filing availability

Unavailable actual sources do not produce substitute holdings.

1. Open a company and its evidence

Use Evidence links in the company profile. The explorer reads the same applied result as the financial profile; opening it does not fetch another source.

2. Distinguish the two captures

The name, symbol, exchange and CIK association comes from a dated SEC catalogue. Financial facts have a separate companyfacts capture, hash and age. An older mapping does not establish the current traded share identity. A filing date is also separate from the time the application captured the source.

3. Choose a cited filing

Search an accession or filter the form and source bucket. Annual, quarterly and YTD occurrences stay distinct. Citation counts describe records in the captured source, rather than independently verified documents.

4. Inspect every original value

The chosen filing has a complete ledger of linked original occurrences: concept, value string, unit, accounting interval or instant, form, fiscal period and source position. Repeated occurrences and differing contexts are not silently merged. Similar numbers in different periods are separate facts.

5. Open the source with its limits

Companyfacts cites a filing accession. An SEC archive link lets you continue to its documents; the application does not thereby claim to have separately captured and checked their bodies. The companyfacts hash is not an individual filing hash. These inputs do not establish AI model, GPU or energy relationships.

6. Export the applied whole

Export preserves the whole applied evidence and the same original result. Search, filters and the open filing must not silently shorten it. Changing the company, mode or source requires the newly applied result; an obsolete reply must not masquerade as the new selection.

1. Check what you are comparing

Both inputs express Czech koruna per one euro. CNB and ECB nevertheless produce their own daily references with different methods and calendars. Matching currency and date do not establish the same measurement instant or definition. Official-method links are available on the screen.

2. Choose a day or period and apply it

Availability comes from the actual catalogue and coverage. The public view automatically prepares a bounded common CNB/ECB period and reads both sources; Apply a custom interval or preset. An unfinished change must not relabel an earlier applied result. A reply for an obsolete selection or account is denied.

3. Read the complete coverage

Chart and table use one preserved result containing both original series. A day with only one source remains visible without a calculated difference. Missing values are neither carried from a previous day nor filled with an average. A weekend or empty day is not a new fixing.

4. Distinguish a difference from an error

The signed difference is CNB minus ECB in koruna per euro. The relative difference uses ECB as its denominator; its percentage has its own declared rounding. A difference alone proves neither an error, an executable trading profit nor a better actual currency conversion. Fees, spread and trade execution are outside this model.

5. Inspect a specific day’s origins

Selecting a day links the chart, table and both source details. Check the original value string, unit, date, method, revision, source hash and actual capture clocks. A central bank’s general publication schedule does not prove a specific observation’s exact publication time. Unknown clocks remain unknown. If the recorded clocks conflict, the original stays in the inspector with a blocking reason; neither the chart nor the difference uses it.

6. Download the original whole result

JSON and CSV preserve the whole applied comparison, its assumptions, both source sides and available evidence. Before download, access to the same result and both sources is renewed. An account, session or permission change must not release formerly opened in-memory data. Viewing or downloading alone does not publish a community contribution.

1. Declare your assumptions

Enter initial capital, first-year annual spending, currency, fixed annual inflation and withdrawal horizon. Assumptions are prefilled and editable, rather than recommendations. Start this more expensive simulation explicitly after reviewing them. Choose withdrawals at the start or end of each year. This is your model without imported market prices, historical calibration, taxes or fees. Changing currency does not perform an FX conversion.

2. Define the return distribution

Supply three personal annual returns and positive integer weights. A weight represents relative sampling frequency, not a historically measured frequency. Years are drawn independently from the same distribution; the model contains no trends, dependence between adjacent years or inflation-return relationship. Your declared assumptions therefore determine the outcome.

3. Run a reproducible simulation

Choose a path count and seed, then run explicitly. The same method, all the same inputs and the same seed reproduce the paths. More paths do not validate the assumptions. Input changes require another run; an older response must not replace a newer draft.

4. Read both the range and a specific path

The annual range summarizes every modeled path, including paths with a previous funding shortfall. Quantiles across different years need not form one actual path. Inspect a selected path to check sampled return, available capital, planned and funded spending and the first unfunded amount.

This simulation permits a partial withdrawal of available money. After the first shortfall, capital remains zero; debt and later income are not modeled. A fully funded final withdrawal with an exactly zero balance is not a shortfall. The separate original withdrawal model uses a different rule: an unfunded withdrawal is not executed.

5. Compare spending sensitivity

Your spending-deviation percentage produces lower, base and higher spending variants. They share the same sampled returns, so differences come only from declared spending. Read each shortfall count with its full path denominator. The fraction applies only within this model; it is neither a validated retirement-success probability nor a recommended safe withdrawal rate.

1. Choose economies and apply the draft

Below the macroeconomic table, the default comparison of available economies is prepared from the same already loaded World Bank batch. Apply comparison confirms your selection of two or three economies, indicator and five or ten common years without another source request.

Editing a draft does not relabel the last applied chart or export. The new query takes effect only after Apply. A refreshed source requires a new Apply so that an old result cannot appear to analyse new inputs.

2. Read the same indicator in common years

Annual CPI inflation is neither the latest monthly inflation reading nor HICP. Real GDP growth is an annual change rather than quarterly growth. Unemployment is the source modeled ILO estimate as a share of the labour force. The euro area is the original EMU aggregate, not an average of selected countries.

Lines share a percent axis. They are neither an index rebased to 100 nor cumulative growth. Select a common year with the mouse or keyboard; the inspector shows each economy, its exact original value, status and footnote. The table retains the complete applied period.

3. Keep gaps and source clocks visible

A null value in an existing source row differs from a row absent altogether. Both break the line and remain empty; zero is a real value. Neither a preceding year nor another economy fills a gap. Source statuses remain intact without an inferred claim of finality.

The year denotes the statistical period. Source database update, application capture and cache expiry have separate clocks. They do not establish the precise publication time of an individual value. A local calculation never relabels expired cache as fresh data.

4. Keep the whole applied result

CSV and JSON share the chart and table result ID. They preserve applied economies, indicator, years, every cell including gaps, the manifest and complete original normalized inputs. Switching table, chart or inspection year removes nothing from the result. Export produces a file; it creates neither an account-saved analysis nor a public share.

Main catalogue and supported fund data

Open Fund and ETF catalogue in the main navigation. An exact product link selects its ID and the public catalogue loads automatically. Profiles for SWDA, EIMI, CSX5, AGGH, IGLT and INRG link to Original facts, NAV, growth and holdings in the main application. Other products retain their whole original catalogue profile; a missing detailed source is never replaced by another fund. Specialist CSPX/IUSA previews remain separate.

2. Load one complete catalogue

The official iShares UK individual catalogue loads on opening. One bulk response supplies the whole list; choosing another product needs no separate source request. A failed refresh preserves the previous applied result.

Coverage means every product returned in this source context. It is not the entire global fund market. The issuer’s ETF group also includes ETP products; a group label or UCITS wording in a name does not prove regulatory status, broker access or suitability.

3. Filter and compare exact products

Search by name, ISIN or ticker and use the available product filters. Apply the draft explicitly. Paging only changes which rows are visible. Select up to six product IDs for the comparison; changing the draft does not silently replace the applied selection.

The catalogue ticker is the issuer’s reported local ticker, not a complete exchange-listing identity. The share-class currency does not establish currency hedging. Open the official product page for further information; the separate CSPX/IUSA comparison retains detailed original share-class facts and listings.

4. Read costs without mixing definitions

TER and ongoing charges are separate source fields. A missing value stays unavailable even if the other field exists. The source’s combined display field does not make the measures interchangeable. Reported percentages are not an individual investor’s total transaction cost or tax outcome.

Capture time and cache expiry describe the downloaded document. They do not supply an unknown effective date for a fee or product classification. NAV and AUM now also appear as dated issuer facts. Their original fields remain in the complete export; this catalogue does not turn them into live tradable prices.

5. Open a product profile and dated performance

Open a product from the applied table. Overview, Performance and Sources use the same captured original and result; changing tabs does not fetch another fund. A standalone product address identifies its exact product ID. Before loading, it supplies only that requested ID; after applying the catalogue, an absent ID stays absent rather than opening a different product.

Annualised returns, cumulative returns and separate twelve-month periods are different groups. The second discrete slot is the previous twelve-month period, not a two-year cumulative return. Each group retains its own reported date; some dates are years old. Calendar slots are normalised only for the source-configured index-fund view and keep their reported ordinal labels.

The issuer reports performance on a NAV basis in its base currency, with reinvested net income and after fees. The profile does not calculate your currency-converted investment return. Net assets have an explicit USD source label and no dedicated reported date; NAV or AUM dates do not fill that gap. Switch the catalogue table to NAV and size, or open the same dated points in a product profile and the applied selection of up to six products. Each value retains the issuer’s exact display and its own date. Click a value to inspect its original numeric lexeme, date, capture and source hash; an old date stays old after a new capture. NAV currency and per-unit mapping are unestablished. AUM (M) means millions of base currency at fund or series level as applicable, but neither its exact currency code nor the per-product level is established. Share-class currency is separate and must not be assigned to these values. No totals, rankings, currency conversion or NAV history are derived. The complete JSON/CSV already contain all original NAV/AUM fields for every product; presentation and point selection do not limit them.

6. Put a reported cost in context

The cost Context Lens uses the same complete captured catalogue. It groups exact issuer product kind, asset class, share-class currency and investor class; the tighter view also matches income treatment. Changing the visible group or TER/OCF measure selects an already calculated context and does not fetch another source. Catalogue search filters do not silently shrink the peer group.

The display requires at least twenty valid values and eighty percent coverage of the whole defined group. These are declared method thresholds. Missing values remain missing; exact zero remains zero. Values outside the method’s 0–100 percent plausibility range remain in the evidence but do not contribute to a distribution. A small ETP group therefore does not receive a dramatic percentile.

Read the valid and excluded counts, lower/equal/higher cost counts and the declared quartiles together. The full membership table identifies every contributing or excluded original product. Products and share classes have equal row weight; this is not an asset-weighted global fund ranking. Matching these structural dimensions does not establish the same benchmark or exposure.

The sample belongs to the captured issuer catalogue. Fee effective dates are not supplied and remain unknown. The position is descriptive context for reported figures, not a historical membership claim, a performance forecast or a preferred investment. Complete exports retain every group and context, including unavailable ones.

7. Inspect and export the same result

Open a product’s original row to inspect the reported fields. The applied query, selected comparison, source capture and full original catalogue belong to one result manifest. Complete JSON and CSV preserve that result rather than exporting only the currently visible page.

1. Load the selected official fund documents

Opening the UCITS comparison reads actual inputs for its default selection of supported funds. Each fund keeps its own document and date. Apply later selection or filter edits with the corresponding control; a missing fund is never substituted.

Each document has its own original capture and cache expiry. A fresh download does not establish a new effective date for every fact. A source error or unapplied edit preserves the previous applied result with its original identity and clocks. No missing source is replaced with a fictional fund.

2. Distinguish the fund share class and its listings

ISIN identifies the share class in this comparison. Its exchange listings can have different tickers and trading currencies while retaining that same ISIN. Several listings are not several separate fund alternatives. The facts table preserves the issuer’s fund and share-class currencies separately from the listing table.

A listing currency does not establish currency hedging or economic FX exposure. The hedging row uses only the explicit reported share-class field and retains its source context. The source currency label GBP alone does not establish a GBp price multiplier. This workspace supplies no executable listing prices, broker availability or currency conversion.

3. Compare costs, income and the reported index

Read the reported total expense ratio as an annual percentage, alongside income use, domicile, UCITS status, product structure and replication methodology. Accumulating and distributing classes handle income differently; this table does not calculate a personal tax outcome or reinvested return. An undated expense field stays undated rather than receiving the document capture date.

The comparison can identify the same reported benchmark name. This is a factual candidate relationship, not proof of identical gross/net index variants, holdings, tracking difference, liquidity or future performance. Neither the shared name nor UCITS status guarantees a suitable or broker-tradable replacement.

4. Inspect original fields and export the whole comparison

Open the source detail to inspect the exact reported value, field path and its own date where supplied. Some explicit source fields are present in the issuer payload without being visible in the issuer’s rendered table; their visibility remains disclosed. Unknown facts stay unknown. Source hashes identify the captured bytes, not independent verification of the fund’s claims.

JSON and CSV retain the entire applied selection, facts, every original listing, scoped original source fragments and the same result manifest. Searching, paging or opening a detail does not shorten them. This local workflow does not create an account Snapshot, a public recommendation or a guarantee of complete market coverage.

1. Open one stock or ETF

Open Stocks or ETFs and select a listing. The supported selection contains thirty US stock listings and ten ETFs, all quoted in USD. A company, a particular listing and an ETF portfolio are separate identities. The current quote and its own capture remain above the history.

The initial three-month history belongs to the existing overview capture. Longer history loads only for your selected listing and period; opening a market overview does not download five years for every instrument.

2. Choose and load a period

Choose one month, three months, six months, one year or five years, then load it. Selecting another period prepares a new request; it does not rename the last displayed result. A source failure leaves the previous applied result available with its original period and times.

The source uses a relative requested range. Read the actual first and last returned observation and point count rather than assuming a complete calendar. A recently listed instrument can have a shorter history. Missing trading days and weekends are not filled with invented prices.

3. Read the daily Close correctly

The series retains Yahoo quote.close with the provider’s split adjustments. It is not Adj Close, dividend-reinvested total return, ETF NAV or an executable quote. Datisto performs no additional split adjustment, currency conversion or interpolation.

A daily timestamp marks the source session start, not a verified closing trade or publication instant. The exchange time zone is shown separately. Completed sessions form the chart; any current session remains identified in the full table and export. Unknown prices or volumes stay missing; an explicit zero remains zero.

4. Move between chart, table and point detail

The chart and complete paginated table refer to one applied result. Search and pagination change what you see, not its source history. Open a point to inspect its exact decimal price, volume, session state and original position. Small graph labels can be rounded; the point detail and export preserve exact strings.

5. Check source and capture

Source details retain the exact listing, requested range, source URI, original-document SHA-256, capture and cache expiry. Returning a cached result does not make its capture current. A checksum identifies the captured bytes; it does not certify complete exchange coverage or independent economic accuracy.

6. Export the whole applied result

JSON and CSV contain the complete applied history and matching result metadata, including rows hidden by search or another table page. They preserve unknown values, current sessions and original precision. Changing a range draft does not change an export until its new result is successfully loaded.

This is a local product-preview history. It does not create an account Snapshot or public share, add dividends or extend the separate existing company-price analysis automatically. Watch and comparison actions in this preview retain their own scope.

7. Compare two to six listings over the same range

Add supported stocks and ETFs to the pinned comparison, then open Compare. Choose a month through five years and load the source range explicitly. Only the selected histories load, one after another, preparing a candidate. Then apply its analytical window. Changing pins, source range or dates prepares a draft; loading alone does not replace the applied chart, its original cohort or period.

A source failure for one selected listing is not permission to drop it and present the others as a complete comparison. Review its reason and retry later; the previous applied result remains available. A mixed selection containing crypto or indices uses the existing separate comparison workflow.

8. Understand the common base of 100

The comparison matches only identical original timestamps with completed, known Close values for every selected listing. The earliest common point inside the applied analytical window with positive prices for all listings establishes the same base of 100. Missing rows are not replaced with a nearby date. Rows before an eligible positive base, current sessions and other exclusions remain explained in the result.

At each admitted time, the index is 100 times the original Close divided by that listing’s base Close. Integer-ratio arithmetic rounds to eight decimal places using half-up rounding. An index of 115 means a price 15% above the chosen base; it is not an annual return or dividend-reinvested total return. A zero after the base remains zero, while zero cannot establish a base.

9. Inspect comparison coverage and complete exports

Read each source’s actual coverage, capture and cache expiry alongside the common window and base prices. The captures are independent and need not be simultaneous. A stale retained history stays labelled as last known. A common window is a property of these captured rows, not proof of a complete exchange calendar or a historical as-of reconstruction.

The complete comparison JSON/CSV preserve every selected input history, aligned original prices and indices, source details and excluded-time reasons under one result identity. Searching, paging and selecting a point do not shorten them. Browser selection recipes retain instrument IDs only; they do not store the period, prices or this result, and reopening a recipe does not start source requests.

10. Apply an analytical window without loading sources again

The requested source range determines which history is captured. The separate From and To fields select an inclusive UTC calendar window within those same captures. Apply that window explicitly to establish a new common opening base and result. Editing dates does not replace the previous chart or export, and applying a different analytical window does not read the source again. Clear bounds to use the whole captured scope.

All selected listings still require exactly matching completed timestamps and a common positive opening point. A narrower window can be unavailable even when each full history has data. A missing listing or date is not replaced with another series or nearest trading day. Complete original histories and exclusions remain in the full export, including records outside the displayed window.

11. Read price changes and observed drawdowns separately

Price change compares each original Close with its own positive opening Close at the common base. Base100, original USD, price change percent and drawdown percent are presentations of one applied analytical result. Changing the chart measure does not acquire data or calculate another result. Exact signed integer fractions accompany decimal outputs; plotted coordinates are approximate.

Observed drawdown compares a Close with the highest admitted Close reached so far inside this selected window. The maximum starts at the window opening, so shortening the window resets its high-water mark. It is the drawdown in the observed common completed-session sample, not an all-time, continuous or intraday maximum loss. A real observed zero after a positive base remains valid and can produce minus100percent.

12. Follow the peak, trough and recovery to their original points

Each listing summary retains its opening and ending original prices, price change and largest observed drawdown. Open its peak, trough or later recovery to inspect the exact original common timestamp, source position and Close. Recovery means the first later admitted Close reaching that selected peak again; absence of such an observation remains unavailable. Equal maxima and worst drawdowns follow the stated deterministic earliest-point policy. If the observed sample never declines, its drawdown is zero and no decline episode is invented.

All calculations use provider split-adjusted quote Close in USD. Dividends, fees, FX, executable transactions and portfolio quantities are outside this comparison. The price change is not dividend-reinvested total return, personal investment performance or a recommendation. Independent capture and expiry clocks remain visible and are not refreshed by applying another analytical window.

13. Compare price-return correlations

The same applied two-to-six-listing comparison includes a Pearson correlation matrix over changes in original USD Close, not over price levels. Every cell uses one identical common sample. Each interval requires the same completed, known start and end timestamps for every selected listing and adjacent original source positions in every history. A known missing, current or null observation is not bridged. Both endpoints must be inside your applied dates, and the starting price must be positive. An observed end zero is a valid minus100% change.

At least 20 accepted intervals are required. A shorter sample or a constant return series leaves the relevant coefficient unavailable, including a constant-series diagonal. Read the accepted count, exclusions and actual interval durations. A weekend interval can be longer than one day; this method does not resample, annualize or certify an expected exchange calendar. An unavailable selected listing is never silently removed.

Open a matrix cell to inspect its paired changes and exact original endpoints. Coefficients near +1 describe similar movements in this sample, near −1 opposite movements and near 0 little linear association; none establishes causation, independence or future diversification. Original prices and exact return fractions remain available. Coefficients use the disclosed finite-precision numerical method; colour alone is approximate. Dividends, personal portfolio returns, FX conversion and an automatically chosen benchmark are outside this calculation. The full JSON/CSV retain the same applied identity, original/performance evidence, matrix, all accepted intervals and exclusions; filters and cell selection do not truncate them.

1. Start with the company identity

The public profile automatically reads actual filings for the selected supported issuer; use Refresh for a later update. Each retains an exact symbol and ten-digit CIK from the fixed registry; a link from the broader SEC catalogue requires both to match. A Yahoo quote symbol alone establishes neither the issuer nor its share class. Opening the page does not load all thirty filing datasets.

A ticker is not a permanent legal-entity identity. A holding-company or issuer change does not automatically merge old and new CIKs. Filing entity name and catalogue capture date are part of the context; a new issuer’s shorter history is not replaced with another similarly named company’s history.

2. Annual periods, quarters and year-to-date totals

Revenue, net income, operating cash flow and EPS retain actual accounting start/end dates. Annual mode keeps 330–400-day intervals from annual filings. Quarterly mode selects original shorter 70–105-day intervals. Exact boundaries take precedence over an assumed calendar quarter.

A six- or nine-month value is year-to-date, rather than an independent second or third quarter. The profile keeps these inputs separate for TTM calculations. Assets, liabilities, equity, cash and share counts are balances at a date; quarterly display does not turn them into three-month flows.

A quarter may be reported in a 10-Q or as a short comparative interval in a 10-K. Its original form and fiscal-period label remain intact. Source fiscal year describes filing context: a fact labelled FY 2027 can have an actual end in calendar 2026.

3. The exact measure and unit

Two revenue definitions remain separate: contract revenue excluding certain taxes and the general Revenues tag. The profile does not join them into one smooth line. Similarly, EPS in USD per share and a share count are not USD monetary amounts. Every row retains its original tag and unit.

Negative net income or cash flow is an original value, rather than missing data, and is not rewritten as zero. A bank’s reported items can have a different economic meaning from similarly named industrial-company aggregates; this screen does not automatically turn them into quality ratings, valuations or recommendations.

Quarterly EPS is not simply added because the diluted share denominator can change. Instant balances and share counts are not summed either. TTM here is available only for supported additive monetary flows.

4. Two auditable TTM methods

The first method adds four original contiguous quarters of the same issuer, concept and unit. They must have no gap, overlap, unknown result or latest-filing conflict. The combined interval represents an accounting year and can contain, for example, 364 or 371 days rather than exactly 365.

When a standalone fourth quarter is missing, the second method uses last annual result + current year-to-date − corresponding prior year-to-date. Current and prior YTD must be comparatives in the same filing. Their boundaries must align with the annual interval, and the annual input cannot be a later amendment filed after the chosen interim disclosure.

Detail shows the formula, each input sign/value, exact periods and filings. The derived result is labelled and never presented as a directly reported quarter. If two valid methods disagree or the inputs admit ambiguous combinations, the conflict remains visible and the value stays unknown.

5. Older values and amended filings

Each summary card has its own date. The last available cash value from 2018 does not become a current balance after a new download. One missing or old tag does not invalidate other established facts, but it does not establish a complete balance sheet either.

The latest established filing date is applied within the same measure, unit and exact period. If the source contains ambiguous latest filings on the same day, detail retains the candidates rather than inventing a winner from ordering or accession number. Filing date is not the exact first-publication time, and the latest amended value is not an investor’s historical knowledge on an earlier day.

6. Growth, margins and profitability with the right baseline

Open Growth and margins and explicitly choose the revenue definition. Net margin divides net income by positive revenue over the exact same interval; operating cash-flow margin divides operating cash flow by that revenue. The two revenue definitions are not substituted. A missing concept, mismatched period or ambiguous filing leaves the calculation unknown.

Year-over-year growth compares the same accounting concept with its corresponding prior-year interval. Fiscal years can contain 52 or 53 weeks, so the method checks both boundaries and duration rather than just a year label. Multiple eligible intervals are ambiguous. A zero or negative prior value does not yield a conventional growth percentage and remains without a percentage.

Detail retains the formula, exact amounts, periods and filings for both inputs. Results use the latest captured disclosures and may include revisions from different filings. They are not a first-known historical backtest or an automatic company rating. Filters and pagination do not shorten the complete applied-result export.

Return on assets (ROA) and equity (ROE) use net income for the actual interval divided by the mean opening and closing balance. The opening balance must be at the exact day before the income interval starts, and the closing balance at its last day. A missing or ambiguous balance is not replaced with a nearby date; a nonpositive mean does not produce a conventional percentage.

Quarterly ROA or ROE describes the result for that reported quarter. It is not multiplied by four or presented as an annual or TTM return. Detail shows all three original inputs and the exact mean denominator. When comparing companies, preserve interval length and actual dates rather than relying only on the ROA or ROE label.

7. Compare companies and their reporting periods

In Compare companies, choose two to six companies and one common metric. Each row retains its issuer identity, unit, source capture and actual reporting start/end dates. Additional companies load only when the selection is applied; the thirty supported symbols are not fetched together.

Each company’s latest period is a separately dated selection. NVIDIA’s fiscal year can end on a different date from JPMorgan’s; a common year label does not make the intervals identical. For an exact reporting interval, choose actual boundaries. Balances compare a date, while flows compare a full interval. A missing corresponding period remains unavailable.

One issuer’s source failure does not switch other rows to demo data. Comparison retains each input’s clock and status; it is not one atomic capture of all companies. Detail and complete export belong to the same applied selection. A bar’s height is not an automatic quality score or investment recommendation.

8. Debt, dividends and repurchases

The capital and distributions view shows original debt items, dividends paid, repurchase payments and share disclosures. Each item retains its exact reporting concept, unit, date and filing. A missing bank tag does not mean that the bank has no debt; similarly, total balance-sheet liabilities are not automatically interest-bearing debt.

Reported LongTermDebt and current/noncurrent component tags can overlap. The profile keeps them separate and does not create total debt by adding every available tag. Debt including leases can have a different scope from debt alone. Inspect the original tag and actual balance-sheet date before comparing.

Dividend and repurchase payments are cash flows over a reporting interval. Annual mode, a standalone quarter and year-to-date totals have different meanings. A six- or nine-month amount is not relabelled as the latest quarter. Repurchase cash does not establish the number of shares retired or the exact change in an investor’s ownership.

A dividend per share in USD/share is not a total USD payment. An older reported per-share dividend does not establish today’s dividend yield, next payment date or reinvested return. Reported common shares are not automatically a listing’s free float or current dilution. Weighted-average shares have reporting intervals and differ from an instant balance. Share-based compensation reported in USD is not automatically a cash payment.

Changing the item or period mode reuses the same captured input. Detail preserves original values and revisions; the complete export contains every admitted period and manifest. Statements and the capital view share the issuer-document acquisition, while their applied results retain their own identities and visible clocks.

9. Detailed statements and distinct accounting concepts

In Detailed statements, switch between income, cash-flow and balance-sheet items. The view contains 22 original concepts: costs, gross and operating profit, research, selling/general/administrative expense, interest, tax, bank noninterest expense, investing and financing cash flow, asset-acquisition payments, depreciation adjustments and five current balance items. Choose a concept and annual, quarterly or cumulative mode, then apply.

Aggregate operating expenses and their components can overlap; adding them does not create another expense. The two cost-of-sales concepts are not substituted. Payments to acquire productive assets have a different scope from property, plant and equipment alone. A bank’s noninterest expense does not substitute for an industrial gross profit. Each card and detail retains the original tag.

Depreciation, amortization and adjustments including accretion remain separate reported concepts. They are not automatically cash payments, EBITDA or free cash flow. A tax benefit and negative cash flow retain their signs. Older reported interest or asset purchases keep their dates; missing bank-specific or custom taxonomy is not zero.

Cumulative mode keeps original year-to-date totals: Q1 spans roughly one quarter, Q2 two and Q3 three. A standalone second or third quarter is excluded from this selection. Instant balances have no cumulative total and remain unavailable in YTD. Chart, detail and complete JSON/CSV use the same applied input; card group, search and pagination do not shorten exports.

10. Applied profile and complete export

Prepare a company, measure and annual/quarterly mode, then apply. Measure or mode changes reuse the same captured batch without another SEC read. Chart, table, comparison and TTM share this result. A separate refresh may obtain a new source or valid cache with its original capture.

Complete JSON and CSV retain annual, quarterly and YTD inputs, revision contexts and derived TTM recipes. Search and page do not shorten the export. Complete first-publication history and broader accounting taxonomies remain further work.

11. Read prices alongside reported results

In Price and company results, choose the original revenue definition, diluted earnings per share or operating cash flow, annual or quarterly periods, and a calendar window. Apply freezes the available selected-company price and filing inputs into one result. The two panels share calendar dates and keep separate units. Table, point detail and complete JSON/CSV refer to that same result. A later source batch does not silently replace it; explicitly apply again to use changed inputs.

Accounting period end and the latest selected filing date answer different questions. A filing date has no intraday publication time here; later revisions can affect the selected original facts. This view therefore does not show exactly what investors knew on a historical trading day. Missing values are gaps, not zero or a nearby future observation. The default three-month price capture can be shorter than the financial history; widening the accounting window does not create older prices. A separate longer price range must first be loaded, as described below.

Yahoo quote Close is adjusted by the provider for splits; it is not dividend-reinvested total return or AdjClose. Its timestamp marks the session start, not the exact closing trade, and a current session remains provisional. Reported EPS can use a different filing share basis, so parallel panels do not establish a P/E ratio. Operating cash flow is not free cash flow. Actual prices are never combined with fictional demo financials.

12. Calculate free cash flow from original contributors

In Free cash flow choose which original capital-payment definition to subtract from reported operating cash flow: productive assets or property, plant and equipment. These scopes remain separate; an unavailable one is never replaced or added to the other. The result is a calculated measure under the displayed Datisto definition, rather than a company-reported standardized SEC fact, FCFF or FCFE. Select annual, quarterly or cumulative YTD periods explicitly. YTD is not presented as a standalone quarter.

Apply freezes both inputs from the same source capture and exact reporting interval. Open the formula or a history row to inspect the original amounts, units, start/end dates and filing candidates. Both contributors must have compatible filing context. Zero capital payments are valid; negative payments are unavailable under this definition. Negative operating cash flow remains negative. When the latest operating period has no compatible capital payment, its result remains unavailable even if an older pair was calculable. A bank with no admitted payment concept does not receive a manufactured industrial cash-flow value.

Refresh obtains a candidate capture through the existing shared SEC cache. The applied history stays tied to its original input until you explicitly Apply again. Source errors do not substitute demo amounts or erase the prior result. Full JSON and CSV preserve both definitions, original source contexts and the result manifest beyond current search or pages. The price comparison view can use the same explicit calculated definition.

13. Compare the calculated cash-flow measure with price

In Price and company results choose Calculated free cash flow, select the productive-asset or PPE payment scope and annual, quarterly or cumulative YTD periods. Choosing FCF obtains its paired source through the existing cache. Apply freezes this source and the selected listing’s captured price history into one result. A newly loaded source, a different definition or a changed date range does not silently replace that result. YTD belongs to the calculated cash-flow choice; the four original reported measures retain annual or quarterly periods.

The two panels use the same calendar window with separate units. Open a calculated point to inspect operating cash flow minus the chosen original payment, both original filing contexts and every candidate. Missing or conflicting latest inputs remain an unavailable attempt; an older successful pair is not carried forward. Under the filing-date axis, the date is the latest input or attempted calculation’s filing date, not a company-reported FCF publication time. It has no intraday ordering against share-price trades.

The full export identifies the applied price analysis separately from its delegated original or FCF calculation. It retains complete captured source records, both FCF definitions and original contributors, while the window ledger contains only the explicitly selected dates. Search and pages do not shorten the full export. With the overview-price choice, the retained three-month price history does not grow when you widen the accounting range; a longer selected price history must be loaded separately. Price remains split-adjusted Close rather than total return; no FCF yield, valuation multiple or historical-knowledge backtest is calculated.

14. Load up to five years of prices before applying

In actual mode choose the captured three-month overview or history for a separately selected range. For the latter choose one month, three months, six months, one year or five years and load it explicitly. The new price capture is a candidate. It does not calculate or replace the displayed analysis until you Apply with the chosen financial measure and calendar window.

Changing the requested range requires a matching loaded candidate before it can be applied. New sources, cancelled or failed reads and old asynchronous replies leave the applied result attached to its original inputs. The original overview-price choice retains its own shorter source capture.

Read actual first and last price observations and the complete point count. A five-year request is not a guarantee of a complete exchange calendar. All admitted daily points remain in the original input; calendar filtering only selects the displayed window. Price and financial panels keep their own units and dates without filling a missing price, turning a quarterly flow into daily values or inferring intraday disclosure order.

The same longer price input can accompany an original reported measure or either calculated FCF definition. FCF still delegates the existing original operating-flow and capital-payment selection; a missing latest pair is not replaced with an older good value. Full JSON/CSV retain every captured price, the original financial sources and FCF contributors beyond the displayed window, with separate source clocks and one applied analysis identity.

15. Full SEC catalogue and original result

Open the full SEC catalogue, which loads automatically. Find the exact symbol, name, exchange and CIK row and choose Open statements. Identity checks and the actual statement read happen directly. Many catalogue rows do not imply available accounting sources for every issuer.

The broader profile uses ten original concepts, annual/quarterly periods and auditable TTM under the same rules as existing profiles. Detailed statements, capital and other specialized views remain in the original thirty-issuer selection. A missing concept, unsupported taxonomy or failed read is not zero.

Pin two to six distinct issuers from the same captured catalogue. Equal symbols with different CIKs stay separate; share classes sharing one CIK are not two companies. Every selected member stays in the comparison even when its statements are unavailable. Changing the measure reuses the captured inputs; refreshing them is a separate explicit action.

1. The fund and its exchange listing

IWM is an equity fund tracking the Russell 2000; TLT holds long-dated US Treasury bonds. The issuer profile describes the fund, while a quote belongs to its particular US listing. A ticker alone is not a universal security identity. This view covers two fixed funds, rather than the entire ETF market.

Exchange price and NAV per share may differ. Each has its own source, time and method. Similar values from different dates are not joined into an exact premium or discount. Quote currency and NAV valuation currency also do not establish currency hedging or the fund’s complete currency exposure.

2. Holdings have their own date

Positions come from the issuer’s direct CSV and include every loaded row, including cash and derivatives. IWM can contain missing tickers; bond fund TLT distinguishes issues through CUSIP or ISIN. A shared Treasury Bond name is not a reason to merge distinct bonds.

The holdings effective date remains separate from download time. Refreshing an older file does not create today’s portfolio. CSV row count may differ from an issuer’s separate security count because cash, futures and definitions can be counted differently. The profile does not promise a historical holdings archive.

3. Weight, quantity and exposure

Sector, country, currency and asset-class groups sum original weights across the same complete file. Search and pagination do not change their basis. Rounded or missing weights are never forced to 100%. Negative cash stays negative and an unknown country stays unknown.

ETF weight is not the fund’s ownership percentage of an underlying company. Market value, futures notional exposure, share quantity and bond par value are different measures. A bond price unit is not automatically dollars per item. A custom price-times-quantity calculation does not replace the issuer’s valuation method.

4. Fees and performance

Expense ratio is the issuer-reported annual expense ratio, rather than a one-off trading fee. It does not establish broker costs, spread or tax. A missing fee effective date remains unknown. Distribution frequency does not supply a payment amount or guarantee future distributions.

The issuer’s performance table has its own as-of date and return definition. One-year and longer annualized figures are not added as simple price changes. Listing price history is a separate series and does not replace return with reinvested distributions. Past performance is not a forecast.

5. One applied profile and a complete export

Choose a fund and allocation, then apply the change. Detail, chart and table use one captured batch. Up to four positions can be compared on their properties; this is not a recommendation to buy them individually. On first load, an available matching quote for the same default identity is attached automatically. Accept later quote updates explicitly so an already applied result does not change silently.

JSON and CSV preserve the complete normalized result and source evidence, rather than just the visible page or largest positions. The local preview does not establish a complete historical archive, personal portfolio fund look-through or public launch.

1. Start with the statistics date

The network summary has its own source added date and a separate response capture. A new read can return an older statistic. The view therefore shows its original date and age at capture; the 48-hour boundary is a local older-input label policy, not a provider promise.

Node rankings have a different capture and supply no as-of date. The network summary date is never copied onto them. Even nearby receipt times do not create a shared current snapshot of the whole network.

2. Public capacity is not an immediately spendable balance

Node count, channel count and total capacity describe the network visible to the provider. Capacity is in satoshis; BTC conversion uses exactly 100,000,000 satoshis per bitcoin. Mean and median capacity are distinct original measures, not two interchangeable estimates of a typical value.

A publicly announced channel reveals neither every private channel, the balance split by direction, available inbound/outbound liquidity nor success of a particular payment route. This screen sends no payment, tests no route and connects to neither a wallet nor a remote node.

3. Node capacity counts channel endpoints

A channel connects two nodes, so its capacity may be included at both endpoints. Summed node capacities are not unique network capacity. This sum is never turned into a global capacity share or a network ownership percentage.

Share bars use only the sum of the returned list for that same metric. They describe that bounded ranking. Omitted nodes do not have zero capacity or zero channels; they are simply absent from this response.

4. Identity and missing values stay explicit

Nodes are matched by their exact public key. An alias is source metadata, not a verified person or independent operator. Similar names are never merged into a supposed owner. If aliases differ between lists, the original text from each list remains available.

A capacity list may lack channel counts, and a channel list may lack capacities. A value is joined only through an established exact-key match in the other returned list; otherwise it is unknown. The separate capacity and channel endpoints currently return 100 rows each. The summary endpoint supplied only six, so the view reads both complete available lists. Missing rows are never invented.

5. One applied view and the complete result

Prepare and apply a metric change. Applying a ranking reuses the captured batch without another network read. A separate refresh may obtain a new response or the original fresh cache. Search, page and comparison change presentation only and never recalculate the share denominator.

Complete JSON and CSV contain both normalized source lists, the original-dated network statistics, every returned node and the common manifest. A local result establishes neither public data rights, decentralization nor payment availability of the network.

1. Choose a block by height or hash

The view starts with the provider’s ten latest blocks. Select a block or enter its height or full hash, then apply the draft. Height is a position in the chain; a hash identifies a particular block. A reorganization can map the same height to another hash.

Block facts, chain status and a transaction page have their own sources and clocks. Block time comes from its header, while response capture records when Datisto received the data. The latest displayed block is not an independently verified tip of the entire network.

2. Keep pagination inside the same block

One source page contains up to 25 transactions. The next page keeps the selected hash; a new block must not automatically replace your choice during browsing. The block’s total transaction count and the currently loaded row count are different quantities.

Changing a block or source page creates a new applied result. Selecting a transaction, local search and sorting only change how that batch is viewed. This does not crawl blockchain history or automatically download every transaction.

3. The map represents virtual size

A transaction’s map area follows its virtual size in vB: weight divided by four and rounded up. It does not represent the value of bitcoin transferred. Weight in WU, serialized size in bytes and virtual size are distinct measures.

A fee is an amount of satoshis; a fee rate is satoshis per vB. Its exact numerator and denominator remain in details and exports, while the displayed decimal is shortened. Fees summed over the loaded page are not the fees of every transaction in the block.

4. Inputs, outputs and coinbase have different roles

Details show each original previous-output reference, its known value and the new outputs. Total output value is not automatically economic trade volume: it may include change and several recipients. An address or script type does not establish an actual owner.

Coinbase is the block’s special transaction. Its absent previous output is not an error, and its outputs are neither an ordinary transfer between two users nor a miner’s net profit. Unknown previous-output values remain unknown. The normalized detail omits complete signatures, witness data and full scripts; it is not a cryptographic transaction verifier.

5. Source status and exports belong to the captured result

Best-chain status is a provider observation with its own capture. A block outside that chain or an unknown status stays explicitly labelled. A negative status does not become confirmation merely because another older response still labels a transaction confirmed. This view verifies neither a Merkle proof nor every consensus rule.

Complete JSON and CSV contain the applied query, source contexts, normalized block facts, the entire loaded transaction page, its inputs and outputs, and the common manifest. They do not export every transaction in the block.

1. Block heights define the epoch

Bitcoin mainnet retargets difficulty every 2,016 blocks. The source supplies the next retarget height and remaining block count. Their difference identifies the height underlying this particular estimate; it is not another independently fetched chain tip. The epoch starts at the next boundary minus 2,016.

Progress is that height minus the epoch start, divided by 2,016. The first epoch block therefore has zero offset. We retain exact counts and denominator alongside the source percentage; no extra block is added to make the indicator appear fuller.

2. An estimated change is not confirmed difficulty

difficultyChange is the estimated percentage for the next change. previousRetarget describes the preceding change and stays separate from the forecast. The provider estimates future pace from block times, using a sliding window early in the epoch. Protocol adjustment bounds constrain the result, but future blocks and their times remain unknown.

The next retarget date is a provider estimate in Unix milliseconds. previousTime is instead the epoch start time in Unix seconds. Mean intervals and remaining duration are milliseconds. Response capture is a separate clock, neither a block timestamp nor independent node verification.

3. Your scenario uses an explicit assumption

Enter assumed seconds per future block, then apply the draft. Your projection multiplies remaining blocks by this interval and adds the duration to its explicitly stated anchor. It does not alter the provider estimate, network difficulty or original input. This is a simple timing alternative, not a probability or an exact forecast.

Editing a scenario reuses the same captured batch without another network read. A separate refresh obtains a new batch or the original fresh cache. Charts, tables, details and exports always correspond to the applied scenario and manifest.

4. History and hashrate retain their own scope

The monthly response has an independent capture. It contains current difficulty and a network hashrate estimate over the last 1,008 blocks; this is not a fixed calendar week. Original H/s are retained, while shortened EH/s are presentation only. A zero error fallback remains unknown.

Hashrate history retains source daily timestamps; the latest day may still evolve. This endpoint groups historical difficulty records by day and averages them, so they are not guaranteed unchanged individual block headers. Their adjustment is a multiplier: for example, 1.05 means approximately +5%, not +1.05%. Small differences from another estimate are never silently reconciled.

5. One analysis, two source contexts

The retarget view and monthly history have separate availability, capture, cache expiry and SHA-256. Failure in one part does not fill its values from another part or with zeros. Nearby capture times do not establish an atomic snapshot. Original age remains visible after cache expiry.

Complete JSON and CSV contain both normalized inputs, every historical row, your scenario and the manifest. A selected detail or current table never shortens them. A local result establishes neither public data rights, guaranteed mining income nor a future Bitcoin price.

1. Start with the period and source

Actual mode uses public mempool.space statistics for Bitcoin mainnet. Choose a week or a month, then apply the draft. The source counts non-stale blocks by their block timestamps within its rolling window; a month is the source calendar month, not a fixed 30 days. Exact computation bounds and block heights are not supplied in this response.

Capture time records when Datisto received the response. It is neither the latest block timestamp nor confirmation that the provider’s whole index is current.

2. Shares retain the whole source denominator

Share is the pool block count divided by the total block count reported in the JSON response. Exact numerator and denominator remain in details and exports; the percentage is a rounded display. Search, pagination and comparison selection do not change this denominator. The provider X-total-count header covers a different scope and is not used.

Unknown pool identity 0 stays in the result. If row counts disagree with the reported total, the discrepancy remains visible; shares are never silently rescaled to 100%. An empty block here contains only the coinbase transaction, rather than zero transactions.

3. A pool identity is not ownership of every machine

Grouping by poolUniqueId joins only identical identities in the provider catalogue. Source-row mode retains its database poolId values. A name or similar spelling does not establish common ownership. Details retain original rows, both identities and their counts.

Attributed block share describes results within a particular window. It does not by itself establish physical ownership of hashpower, the number of independent operators or whole-network decentralization. Short windows can be sensitive to chance and are not automatically future shares.

4. Estimated network hashpower has its own window

The provider supplies three separate network estimates in H/s: based on recent block counts over 24 hours, three days and one week. Those contexts do not become monthly merely because monthly pool ranking is selected above. They derive from Bitcoin Core getnetworkhashps estimates, rather than meters on every machine.

A provider zero estimate may indicate its own failure, so it remains unknown here. This batch does not establish a pool’s hashpower, actual payouts, pool fees or profitability. Audit fields avgMatchRate and avgFeeDelta are outside this normalized view.

5. One applied result for chart and export

A period or grouping edit stays a draft until applied. Chart, table, detail and comparison retain the same applied batch and clocks. A late response from an older request must not replace a newer choice. Expired local cache remains visibly old; serving another response is not a fresh source capture.

Complete JSON and CSV include original normalized rows, the whole ranking and a shared manifest carrying the query, methodology and source proof. Search and the current page never shorten the export. SHA-256 identifies the original source response; the local result identity identifies a particular computation. Local preview does not establish public data rights.

1. Two layers, two sources

The execution source supplies finalized blocks. The consensus source supplies a finalized Beacon header and selected validator state pinned to its state root. Each layer has its own capture, original proofs and context. They do not form an automatically synchronized global snapshot.

Chain ID and known mainnet genesis identities are checked. Finalized is a provider-reported property; this view does not synchronize or validate a complete node. Unavailability in one layer does not overwrite the other layer with zeros.

2. What window burn means

Choose six or twelve blocks. The original execution base fee in wei per gas is multiplied by that block’s actual gas used. The sum is burn within this contiguous selected batch; chart, table and detail use the same original blocks.

This calculation excludes blob burn, priority fees, MEV and newly issued ETH. It establishes neither lifetime burn, net issuance nor a validator’s total income. Gas used, gas limit, wei, Gwei and ETH are distinct units; their conversion does not use a dollar price of ETH.

3. Selected validators are not global staking

Enter at most four distinct indices. The source may return a different order and omit an unknown index, so exact identity is matched and missing output remains a gap. Balance and effective balance are separate original quantities in Gwei.

A sum covers only the returned selection. It does not multiply an assumed 32 ETH per validator and is not global locked ETH, total validator count or future yield. Status, slashed flag and activation or exit epochs remain properties of that particular validator.

4. Aggregates and missing values

ETH supply, total issuance, global staking, queues and yield require their own suitable source and methodology. When connected inputs do not establish a value, it remains unknown or unconnected. Zero appears only when the source actually establishes zero.

When a metric is missing, other evidenced metrics keep their own units and clocks. A blank is not zero, and an unavailable source is not replaced by an invented network.

5. Applied result and complete export

Window or index edits start as a draft. Apply them explicitly; until then, charts, tables, details and exports retain the previous applied query and original clocks. A late old response must not replace the current request.

Complete JSON and CSV belong to the same manifest and include both original contexts, every input block and returned or missing indices. Presentation page and selected detail do not shorten a complete export. Serving a response again does not refresh previously captured data.

1. Start with chain height and the right clock

Actual mode takes the Bitcoin mainnet tip height from the existing mempool.space network connection. Height identifies a zero-based block index; it is not a count of all blocks. The displayed source and capture time belong to that response. Other network and sentiment responses have separate clocks and are not substituted into the halving calculation.

The height endpoint supplies neither the timestamp nor the hash of that block. Reading the response therefore does not establish when the block was mined. Using capture time as an estimation anchor or entering another anchor is an explicit scenario assumption.

2. Distinguish the current block from the first block of an era

The schedule uses Bitcoin mainnet parameters: a 210,000-block interval and an initial 50 BTC subsidy. Integer division of height by the interval identifies the era. The block immediately before a boundary uses the previous subsidy; the block exactly at the boundary already uses the new one. Distance to the next boundary is measured from the input height; no future block is treated as confirmed now.

The calculation uses integer satoshis with integer halving rather than rounded decimal BTC. Fractional satoshis are discarded in later eras until the subsidy reaches a known zero. The table also includes this terminal zero era. Zero subsidy is a specific rule-based result; an unknown source height remains a different state. The method links to the pinned Bitcoin Core v30.0 implementation.

3. Apply timing assumptions as a scenario

For a timing outlook, choose a starting height, time anchor and assumed seconds per block. Faster and slower assumptions form a range from the supplied inputs. The range is neither a confidence interval, a probability nor a promised date. Actual future block timing may differ; an assumed average does not establish the time of any particular future block.

Form edits begin as a draft. Only deliberately applying valid inputs recalculates the result. Until then, chart, table, selected era and export retain the previous applied query. Invalid input does not overwrite the valid result. Refreshing network height is separate and may return the same captured input from shared cache.

4. Connect chart, table and detail without inventing history

Chart and table describe the same subsidy rules by block era. Select an era to inspect its starting and ending height, subsidy per block and theoretical issuance. Unknown actual historical times are not filled by reverse estimation. A future boundary date follows only from the applied anchor and timing scenario rather than a confirmed block that does not yet exist.

Theoretical scheduled issuance does not measure circulating or spendable supply. It totals rule-permitted subsidy rather than establishing what miners actually claimed, which coins were lost or what can be spent today. Genesis has a special spendability restriction and a theoretical total must not automatically become an available balance. Transaction fees are not new issuance and this calculation does not include them in block subsidy.

5. Export the complete applied result with provenance

Complete JSON and CSV belong to the same applied query and manifest as the chart, table and detail. They retain the era schedule, exact integer values, timing assumptions and source context. Selecting one era in detail does not change the scope of a labelled complete export. When the form has changed, apply the new inputs first if you want to export their result.

1. Choose the original market and amount

Actual mode reads the public Coinmate BTC/CZK or BTC/EUR book; euro also includes a separate Coinbase Exchange BTC/EUR book. These are orders on named venues, rather than CNB fixings, an aggregate bitcoin reference or a binding retail exchange offer. Each source retains its own timestamp and capture. Two captures do not establish a shared atomic instant.

The amount means BTC quantity traded in the book. For a purchase it is not a promised net amount after a wallet transfer; for a sale a transfer fee increases total BTC spent. Prepare a change of direction, currency, quantity or fees, then apply it deliberately. The previous result keeps its own query until then.

2. Read the price for the whole amount

A purchase consumes asks from the lowest price; a sale consumes bids from the highest price. Each level contributes only its available quantity. Detail preserves price, consumed BTC and the exact product for every consumed level. Best bid, best ask and their spread describe the top of the book; they do not alone determine the price of a larger order.

The calculation uses at most one hundred best levels per side and reports the original count when established by the response. Insufficient retained depth shows filled and remaining BTC and partial value, while leaving complete price and rank unavailable. Purchase consideration rounds upward to cents, sale consideration downward; inputs retain the original unrounded sum.

3. Separate unknown fees from your assumptions

Manual mode lets you explicitly assume a percentage trading fee, a fixed fee in the selected currency and a transfer fee in BTC. A blank field stays unknown; enter zero only as your own assumption. The same assumptions apply to every compared venue and never become verified provider terms. The percentage fee uses rounded trade consideration and rounds to two decimal places with half-up rounding.

4. Compare the complete cash flow

For a purchase the model adds trading and fixed fiat fees to consideration and subtracts BTC transfer cost from acquired BTC. The complete unit price is fiat paid divided by net BTC received. For a sale it subtracts both fiat fees from proceeds and adds BTC transfer cost to BTC spent. The complete unit value then means net proceeds per total BTC spent.

Ranking requires full quantity coverage and every required fee known in one currency. Purchases rank lower unit cost first; sales rank higher net unit proceeds first. Negative proceeds after high fees are not rewritten as zero. A purchase with zero or negative net quantity has no valid unit price. Chart, table, detail and up to four comparison cards use the same applied result.

5. Preserve the inputs and their limits

Refresh is explicit; this screen does not promise continuously executable quotes. Books retain their capture, source time and genuine original-response hash. A short server cache limits reads but does not establish an offer-validity interval. Binding quote validity remains unknown. An old result stays readable with its own timestamp until a new applied result is available.

Complete JSON and CSV preserve the shared query, all source books, consumed levels, fee assumptions and manifest, regardless of local search or selected cards. A recipe checksum is distinct from a cryptographic source hash. Nothing here submits an order, sends money or creates a durable canonical Snapshot.

1. Start with the network, units and supply

The SOL detail shows a market price; the network view reads the public mainnet-beta RPC. Before using network data, it compares the genesis hash with the fixed identity of that network. If verification fails, other sections are not presented as actual Solana state. A price quote and a network sample have different sources and capture times.

Total, circulating and non-circulating supply are originally integer lamport counts. One SOL equals one billion lamports; the view converts exactly to nine decimal places. The original integer is in the tooltip and JSON. The two supply partitions must sum to the total. Circulating supply alone does not establish free float, an unlock schedule or staked volume.

2. Epoch, slot and block mean different things

An epoch has a source number, length in slots and current zero-based index. Absolute slot and block height are separate fields and cannot be interchanged. Displayed progress is index divided by epoch length, rounded to two decimal places. It is neither a count of produced blocks nor a prediction of the epoch end time.

Epoch, supply and vote accounts request finalized commitment. Individual RPC calls can still return different contexts. Independent capture times help identify input age; they do not promise an atomic shared snapshot.

3. Read TPS together with sample duration

Each sample supplies a transaction count, duration in seconds, slot count and ending slot. TPS is the original count divided by the original duration, rounded half-up to six decimal places. Total transactions include votes; the second series uses only a supplied non-voting transaction count. Missing values remain gaps, and a known zero remains zero.

The horizontal axis uses ending slots, not inferred UTC times. Recent performance samples accept no commitment parameter and do not establish finality or user transaction success. Choosing 30 or 60 requests at most that many recent points. The source may return fewer or an empty window; this is not long history or arbitrary backfill.

4. Refresh, inspect and download the same result

An actual read runs on opening and explicit refresh. A short shared cache limits repeated calls, so the button need not issue a new source request. Each method has its own capture and any unavailability reason. Unavailable sections are never replaced with zeros or synthetic data.

The sample chart and table use the same query and manifest. Full source JSON contains that exact bundle: all supplied samples, all vote accounts, epoch, supply and original integer values. A local validator filter does not restrict the export. A result identity or hash alone does not establish a retained historical archive. Preserve the selected limit, source clocks and methodology for further work.

1. Select a specific market

Start with the venue and exact pair: BTCUSDT, ETHUSDT or SOLUSDT on Binance Spot. The base asset and quote USDT have different units. This view is neither an all-exchange price average nor a perpetual contract. A USDT price is not relabelled USD, and changing the pair creates a new result.

The order book and trade tape show actual messages received for the selected market. An outage remains visible; invented orders or trades never replace it.

2. Read both sides in their proper units

A bid is a resting buy offer and an ask a resting sell offer. Price is USDT per unit of the base asset, while quantity counts those base units. The best bid and ask give the spread as ask minus bid. This difference alone does not establish fees, the obtainable price of a larger trade or another exchange price.

The chosen number of levels bounds the current result. The depth chart accumulates quantity outward from the best price using the same offers as the table. An uncovered part of the market is not zero liquidity. Even the bounded initial snapshot does not establish every distant unchanged level of the book. Resting offers can disappear before execution.

3. The tape shows a different event from the book

The tape contains a bounded set of recent trades received during this source session. Each carries its original identifier, price, quantity and execution time. It can be short or empty immediately after connection; the application does not invent missing history. It is not a complete archive of the trading day.

The buyer_is_maker flag states whether the buyer was the resting maker. If so, the aggressive side was selling; otherwise it was buying. That side does not reveal the participant’s identity, other positions, motivation or the future direction of price. A trade and a subsequent book update carry their own clocks, rather than an atomic global tick.

4. Watch continuity and pause state

A continuous state requires an initial snapshot and consecutive numbered updates. Old duplicates are not applied twice. A sequence gap or disconnected stream requires the book to be rebuilt; old values must not be labelled as a currently valid book. Connection state is therefore part of the result.

Pause freezes the displayed result. It does not stop the exchange source or future market changes. Resume explicitly returns to the current state. An inactive or hidden page does not keep issuing display reads, and its old display is not represented as live.

5. Keep the exact displayed result

The JSON belongs to the same bounded query and manifest as the chart, table and tape. It includes original financial strings, selected pair, level count, sequence and separate input clocks and origins. Exporting while paused preserves the frozen result, rather than an unseen newer batch.

1. Save the asset list

On Comparison, choose two to six distinct assets. Open Saved selections in this browser, enter a name and save the current selection. Keep up to eight named sets. Their order and public identifiers are stored, rather than prices, computed points, time windows or account information.

2. Review the selection first

Loading a saved set opens its draft. Review every item and whether its identity appears in the current catalogue. This alone does not establish a current price, available history or rights to particular data. An unavailable item is not omitted. Merely opening the draft does not change the current selection or run a calculation.

3. Explicitly apply the selection and choose a window

Apply the complete selection and choose a window in Comparison. Prices and history are checked separately for every asset. The saved list itself retains neither a finished chart nor earlier prices.

4. Manage your sets

Rename a set, delete it and restore its deletion with the available undo action. Renaming does not change the asset list. Storage failures remain visible; opening the page does not overwrite corrupt data. A change made in another tab may need to be reloaded and reviewed.

5. Distinguish a selection, a result and a saved analysis

Selections remain in this browser and do not synchronize with an account. The last active list is remembered separately from named sets. On return, check the source, time and window of the newly displayed values.

1. Choose the source and period

From Macro, open the Economic calendar. It starts with official US Bureau of Economic Analysis releases, which need no key. Explore a month or week, choose a day and use search or the release-family filter. An empty week means this dataset has no scheduled event in that period; it is not populated with sample events.

The BEA and FRED calendars have separate sources and coverage. FRED requires a configured server-side key; without it that source remains unavailable. Missing actual events are never replaced with fictional ones.

2. Distinguish scheduled time from measured data

BEA supplies an original timestamp with an offset; the screen retains it and also displays UTC. Open an event to inspect its original title and detail. This is a scheduled release, rather than confirmation that it occurred or that a measured value is available. Reference period, revision, consensus, impact and indicator values are not inferred from the schedule.

3. Inspect provenance and repeated entries

The receipt includes the original machine-readable source URL, capture and document hash. These differ from the scheduled event time. The source file-update value may lack a time zone; it then remains original text with an explicitly unknown zone rather than a UTC timestamp.

The same release family at the same instant appears as one event. Original occurrences and positions remain in the receipt and export. Different families scheduled at the same time are not combined. Whole-source coverage is separate from the filtered month or week.

4. Export the same selection

Calendar, list, detail and complete exports use the same loaded result and filters. CSV and JSON are not limited to the current table page. BEA ICS uses the supplied UTC timestamp and identifies a tentative event without inventing a duration. FRED date-only and sample events remain all-day entries with their respective source labels.

1. Choose exact identities and apply

On first opening, crypto comparison prepares BTC and ETH if you have no existing selection. A selection of two to six identities and changes to the 7, 30 or 90-day period load automatically; Refresh comparison repeats the same request. An explicitly cleared list is not refilled. Identity uses the original CoinGecko ID, never a ticker match.

2. Read the actual shared instants

Only exact original Unix-millisecond timestamps with usable prices in every selected series enter the shared result. Times are not rounded to an hour or day, shifted or replaced by a neighbouring observation. Each source may return a different sample or coverage; a sparse or empty intersection is an honest result. This is not a daily closing-price comparison.

3. Understand the common base

Every series begins at 100 at the first shared timestamp with positive prices in all series. Index 100 equals original price divided by that series’ base price times 100, rounded to 8 decimal places with integer arithmetic. The underlying decimal prices remain unchanged. A later zero is valid; a null or missing observation is not zero. This compares price movement, not dividends, fees, staking income or total investment return.

4. Inspect the same result in every view

Switch the graph and exact table, select an original shared point and inspect every original USD price and its base 100 index. Pagination changes the displayed rows only. Whole CSV/JSON preserve the applied selection, complete source inputs and receipts, common points and excluded timestamps; they are not a download of only the current page. Unapplied draft edits retain the previous result with its original identities and period.

1. Choose days and assumptions

In Calculators, open BTC purchases from original prices. Both selected dates are inclusive source days in UTC. The interval is at most 60 completed calendar days. Choose daily steps or every seven days beginning with the first selected day, a positive USD contribution, an explicit fixed nonnegative purchase fee and initial USD cash. An empty fee is not silently zero.

2. Read purchases, fees and carried cash

Each scheduled step adds the selected contribution to carried cash, then purchases the largest supported quantity after its fixed fee, rounding quantity down. A step with no purchase applies no purchase fee. Initial cash plus contributions equals spending plus applied fees plus remaining cash. Days between weekly steps add no contributions.

The chart shows accumulated modeled BTC, while USD amounts remain in the cards and complete schedule. The final cash-threshold column uses that row’s price and fee for one further quantity quantum; it does not forecast the next day. Average price is spending divided by quantity; average acquisition cost also includes fees. Rounded cards use ≈ and exact output strings remain in details and exports.

3. Inspect the actual original close

Open a row’s price Passport to inspect the exact Coinbase BTC/USD daily Close, original OHLCV context, evidence, revision and capture. The modeled purchase instant is the end boundary of the completed UTC candle, not its start, an exact closing trade or a provider publication instant. Different days may come from different actual captures.

Every scheduled day needs its own usable original price. A missing, nonpositive, withheld or unsupported price refuses the whole model; no day is skipped and no previous close is carried forward. The selected original source days between weekly purchases are still preserved as inputs.

4. Keep the model separate from actual holdings

This is a hypothetical dated cash model, not an order, a recorded trade or evidence that a close could have been achieved. It does not calculate a current market value, profit, investment return, currency conversion or tax basis, and does not write portfolio events. The source’s current local development access does not establish public redistribution rights.

1. Find an instrument beyond open holdings

Choose your private portfolio and open Instruments. The list uses exact instrument identifiers in your recorded events and projected positions, lots or sales. It can therefore include an instrument you no longer hold, an acquisition after the selected economic cutoff, or an original that has been reversed. A cash transfer does not become an instrument because its text resembles a ticker.

The recorded identifier is not a verified exchange listing or an automatic market-price assignment. No open position at the displayed cutoff does not by itself prove a closed investment, a closed broker account or zero lifetime return.

2. Read positions and lots in their own units

The detail uses one complete portfolio read and one economic cutoff for every section. Position amounts remain separated by accounting currency; lots retain their owning accounts, quantities and recorded basis. A quantity is not a money amount. An unknown basis stays unknown. The screen does not allocate portfolio cash to an instrument or infer a current market value from its original purchase price.

3. Follow sales and dividends to their originals

A realized sale retains the existing FIFO result and its consumed acquisition lots. After an account transfer, acquisition provenance may belong to another account. Open the original or lot inspector to read that connection, then return to the same selected instrument. Recorded dividend rows retain their gross amount, withholding and net amount from the existing complete-ledger calculation. An empty section means this source contains no corresponding rows; it is not populated with estimates.

4. Keep reversals, future events and provenance visible

Related activity includes the instrument’s original events, both directions of an instrument transfer and reversals linked to those originals. Events after the economic cutoff remain identifiable history rather than current holdings. Trade fees stay inside their original trades. Separate fees, FX conversions, deposits or withdrawals are not guessed into an instrument’s activity. Required acquisition provenance is labelled separately from selected activity.

5. Export the whole detail and return

Whole JSON and CSV include every selected row and required original record, independently of the visible table page. Download rechecks current private access and reproduces the same complete source at the displayed cutoff. If the source changed, explicitly verify the new detail instead of mixing old and new sections. The ledger hash identifies the complete source, not the subset file’s bytes or a restorable backup. Return to Instruments to choose another exact recorded identifier.

1. Open a company annual sheet

The prepared Apple annual sheet loads automatically on first opening; select Microsoft and open its sheet to change issuer. The selection uses that company’s three named original SEC sources in the retained reporting window. A later company choice is a draft until you apply it; the visible result continues to identify its originally applied company. A missing source cannot be replaced by a similarly named share-price series.

The sheet groups original facts for convenient reading. It performs no valuation or ratio calculation. Revenue, net income and assets remain three different accounting measures in original reported USD, rather than a new combined metric or a conversion to your portfolio currency.

2. Read the complete interval beside the amount

Each row is labelled by an actual reporting-end date from the inputs. Revenue and net income describe a duration with a start and an end. Assets describe a balance at one instant. An assets figure is therefore not an annual flow, and the end date alone does not tell you how long a revenue period lasted. Read each cell’s own interval before comparing.

Choose one of the actual reporting-end dates to bring that group’s three original facts into the summary cards, or open a history row there. Older, newer and latest move only between dates present in this result. This changes the view, rather than executing another financial query. Renewing the same result keeps the choice; opening a different result starts from its own latest group. Export and Snapshot still retain every original input, and the selected card period is not a separately saved Snapshot setting.

The grouping does not turn fiscal periods into calendar years or reconstruct a complete statement. Apple and Microsoft can close their years on different days. The source filing’s fiscal-year field can describe a later filing that includes an older comparative amount. Neither that field nor a year on an axis replaces the original accounting dates.

3. Keep every filing and source clock attached

Open an original amount for its Value Passport. Check the exact amount, USD unit, period, company CIK, US-GAAP concept, accession, filing form and filed day. Different cells grouped under one end date can legitimately come from different filings. Their placement beside each other does not certify a shared filing or make them a suitable pair for an unimplemented calculation.

A date-only filed day does not establish the exact time of publication. The time Datisto captured a source and the time this query ran are separate clocks. Latest uses eligible stored revisions, which can include comparative facts captured in a later filing; this is not a history of what investors first knew. Observed-by-Datisto instead uses the explicit platform knowledge cutoff.

4. Distinguish zero, missing and an ambiguous cell

A reported zero is an amount. Missing source data, an absent counterpart and a quality-withheld amount remain unavailable, with their own explanations. An absent counterpart has no invented observation or Passport. The sheet does not carry an older amount into a newer end date or add a year merely to make the table look complete.

If the admitted original inputs contain more than one candidate for the same measure and end date, the cell must expose that ambiguity rather than select one by display order. Earlier revisions are not additional current alternatives merely because they remain in storage. Inspect the original candidates and their contexts before drawing a conclusion.

5. Download the whole result and retain an original

CSV contains all original rows with their accounting-context columns; JSON retains the typed result, selected inputs, source evidence and grouping document. Presentation or paging does not reduce that full cohort. Each download verifies current access again. The file’s original source clocks remain distinct from the later access-verification time.

1. Identify the captured selection

The screener uses a single actual captured CoinGecko selection of up to250 assets in USD; read its actual member count. Older saved originals retain their twenty-member source. It does not cover every coin, exchange, stock or fund. An asset absent from this selection has not been rejected by your conditions. Provider updates and Datisto capture times have different meanings; neither establishes the first publication time or a guaranteed live quote.

Price is an aggregate reference, rather than an executable offer on a particular market. Market cap is in USD and the original rolling volume covers twenty-four hours. Selecting a seven-day change does not turn that volume into seven-day volume. Percentage changes are source-reported values, not a newly calculated total return.

2. Enter exact numeric rules

Choose price, market cap, twenty-four-hour volume, source rank or the source-reported twenty-four-hour, seven-day or thirty-day percentage change. Up to twelve conditions are combined with AND: every condition must hold. Greater than and at least differ at the boundary; equality compares the exact decimal value. Use a decimal point and no grouping separators. A blank input is not zero.

A hypothetical market-cap threshold of 100000000000 means one hundred billion USD in this field. It is an illustrative personal condition, not a valuation recommendation or a claim about how many assets currently qualify. Select the sorting field and ascending or descending order separately; sorting changes order, not membership.

3. Apply and inspect the actual outcome

Editing conditions prepares a draft. Apply explicitly evaluates the selected captured inputs. An already displayed result remains tied to its applied conditions while you edit; it is not silently relabelled as the unsent draft. Read the applied rules next to the match counts before saving or using the rows.

Matches, excluded assets and unknown outcomes are separate. Missing required values are not treated as zero or as a match. Open the explanation for other assets to see which conditions failed or which fields were unavailable. A successfully evaluated selection with no matches is valid; an unavailable source does not prove that no asset qualified.

4. Check the Passport and export scope

The table contains authoritative matches in applied order. The Asset Passport opens the underlying original point and its provenance. Financial numbers remain exact source decimal strings even when a localized display groups digits. Missing values remain missing.

CSV contains the same ordered matches as the table. JSON additionally retains all original source inputs and the authoritative conditions/outcome document. The full source input count is not the match count. Pages only divide the displayed matches; CSV retains all ordered matches. Export, original reopening and later reads require the current source permissions; saving a Snapshot does not grant permanent access beyond them.

5. Save and return to the original

Review the proposed name and exact result, then confirm Save. The original Snapshot binds its rules, source selection, outcome, ordering and actual recorded save time. In Saved screeners, search by name, identity or condition and sort by current view modification, name or source window. The library shows current mutable view metadata; the original link independently verifies the immutable saved recipe and result.

Opening an original does not recalculate it with current prices. Edit as a new screener transfers only the rules into another draft. Inspect its newly selected source and use Apply explicitly. The old original remains a record of its own captured selection, not a promise that the same assets qualify later.

6. Add verified matches to your own list

Add matches to a list opens a separate review. Choose a new list name or an existing own list and check the specific matches to copy. Up to100 matches start selected; when there are more, none are preselected. Search changes only the displayed matches and preserves your selection. Clear the selection or adjust individual checkboxes; no first100 is copied automatically. Before confirming, review selected versus all matches, actual additions, assets already present and the destination membership after saving. Existing members retain their order, and missing selected entities append in original result order. The complete source result stays unchanged. A list can contain at most100 distinct assets: if the resulting union exceeds this capacity, uncheck items or choose another destination. When every selected match is already present, the no-change review sends no write.

This action copies entity membership once. It does not buy assets, edit portfolio quantities, remove former members or automatically rerun the screener. Future condition monitoring is a separate explicit rule; an ordinary watchlist is not an automatically synchronized screener result.

7. Save dashboard settings for crypto, stocks or ETFs

The dashboard market screener is a separate workspace with a captured selection of up to250 cryptocurrencies,30 stocks and10 ETFs. After Apply conditions, use Save applied settings, choose a name, review the exact conditions and confirm. My analyses and Return to work reopen the group, filters and ordering as a draft. Confirm Apply conditions to evaluate the available current batch. This saves settings rather than the complete original Snapshot described above. Review a changed saved version again; a removed entry cannot be adopted, and expired verification requires account re-verification first. Renewing verification of the same version preserves a valid review.

The eight browser presets remain local and are not copied automatically. Watchlist only uses the current local browser watchlist; saved settings do not retain its members. Rename and remove account settings in My analyses. An unconfirmed save can retain the exact request across a reload of the same tab when session storage is available. Restoration does not send it: verify the account and explicitly retry. Closing the tab does not guarantee this recovery.

1. Define the personal model before entering amounts

This tool models one unleveraged cash purchase followed by a sale at your assumed stop execution price, in one common currency. Select that variant explicitly. Enter the cash amount you are willing to use as the model loss budget, a positive entry price per unit and a nonnegative stop price below entry. An empty price is not zero; an explicit stop price of zero is a supported complete-price-loss hypothesis.

The calculator supplies no recommended percentage of your savings or portfolio. It has no account capital input and does not check whether you can afford the resulting purchase. The displayed entry notional is quantity times entry price, an informative amount rather than a cash availability decision. Currency labels do not convert amounts: all prices, budget and fees must already use the same currency.

2. Read the exact quantity step and price distance

Choose whole units or the explicit fractional model step of 0.000000000000000001 units. The price distance is entry price minus assumed stop price. The gross quantity is the budget divided by that distance and rounded down to the selected step. It excludes transaction fees. The tool uses an exact bounded integer quotient for this floor, so presentation rounding cannot increase the admitted quantity past the budget.

A fractional step is a mathematical modelling choice, not evidence that a broker or instrument supports that quantity. For each quantity, inspect its entry notional, price loss and unused budget. The next-step amount shows why one more selected quantity step would exceed the same budget. This comparison belongs to the stated assumptions and step; it does not rank instruments or determine the best trade.

3. Keep an unknown fee separate from an explicit zero

The fee field is one own fixed cash amount for the hypothetical completed purchase and sale together. It is neither a percentage nor a fee per unit, provider tariff, spread or slippage estimate. Leave the fee unknown if you do not know that assumption. Gross quantities remain available, but fee-inclusive quantity and model loss remain unavailable. Marking the fee known requires an actual amount; enter 0 explicitly for your zero-fee hypothesis.

With a known fee, subtract it from the budget before flooring the remaining amount over the price distance. For an illustrative own model with budget 100, entry 10, stop 8, whole units and a fee of 10, the gross bound is 50 units and the fee-inclusive bound is 45. Its price loss is 90 and total conditional model loss is 100. One more unit would make 102. These are hypothetical arithmetic inputs, not a market price, suggested budget or transaction.

4. Interpret zero quantity as no hypothetical trade

If the fee consumes the budget, or the remaining budget cannot support even one selected step, the result reports no positive quantity. This model then makes no hypothetical trade: applied fee and model loss are zero. The entered fee remains visible in the original assumptions; it is not silently changed to zero. The next positive step includes that entered fee, showing why a positive trade does not fit.

Do not confuse no-positive-quantity with a missing calculation or rejected input. A valid small budget can produce a genuine zero quantity at the chosen step. A stop at or above entry, a nonpositive budget or malformed amount is rejected instead. Changing any input conceals the previous result until you calculate again, so an older quantity cannot appear to belong to a new stop price.

5. Preserve the exact original and separate its clock

Use the exact detail to compare original entered strings with canonical working values. Inputs such as 10.00 retain their original spelling in the account or input-only JSON, even when working arithmetic uses 10. Calculation time on this device describes this new local computation. It is not a source observation, order timestamp or promised original calculation time when you reopen stored assumptions.

6. Understand the conditional stop-price limitation

The SEC investor bulletin on stop, stop-limit and trailing stop orders, updated 18 August 2026, explains that a stop trigger can become a market order and its execution price can differ from the stop price. A stop-limit order can remain unfilled. This calculator does not simulate liquidity, gaps, broker rules or execution. Actual loss can therefore exceed the displayed conditional budget.

Use the result to inspect your own explicit arithmetic assumptions. It covers neither leveraged or short products nor taxes, currency conversion, partial fills, account cash or ongoing income. No button sends a broker order. If a required condition is unavailable, keep it unavailable and continue reviewing the inputs instead of reading a substitute quantity as an accepted loss limit.

1. Choose the explicit cash-long model

Choose the named unleveraged cash-long variant, a common currency and a positive quantity. Enter a positive purchase price and your hypothetical sale price. Quantity is the same for entry and exit, so this is one complete hypothetical purchase and sale, rather than a sale of part of a portfolio lot. A sale price of zero is supported as a total price-loss assumption. The calculator does not know the instrument, fetch a quote, verify an execution or read a portfolio position. Supply your own assumptions rather than treating a book value or the price of another instrument as a market quote.

All prices, fee amounts and outputs use the one selected currency. Changing its label does not convert the entered values. The quantity represents units of the asset and is independent of the cash amount. Enter decimals with a point, without exponent notation or thousands separators. The bounded input format accepts up to 24 integer and 18 fractional digits. The model describes cash price differences; short positions, futures contract multipliers, funding, margin and liquidation require different product-specific rules.

2. Read gross proceeds and the price difference

Gross purchase value is quantity multiplied by purchase price. Gross sale value is the same quantity multiplied by sale price. Their difference is gross PnL before transaction fees. For explicitly hypothetical quantity 2, purchase price 100 and sale price 120, these amounts are 200,240 and 40 in the chosen currency. A lower sale price gives a negative gross difference. This is a scenario result, not a statement that a real trade has closed, a broker confirmation or a recorded portfolio gain.

The result and its breakdown come from the same accepted input bundle. Each edit hides the old result until you explicitly calculate again. Exact cash products and differences preserve tiny supported values rather than turning them into zero through display rounding. A long displayed value marked ≈ is presentation rounding; read the original financial strings and working outputs in the detail when precision matters. An unsupported input or numerical result is refused instead of silently replaced with an estimated amount.

3. Keep unknown fees separate from known zero

Purchase and sale fees are separate cash amounts for the entire respective transaction, in the selected currency. They are not a percentage, a fee per share or a provider tariff. The prefilled amounts are editable model assumptions; mark a value unknown when you cannot establish it for your scenario. To model a known zero fee, explicitly enable that fee and enter 0. A blank enabled fee is incomplete and cannot be treated as zero. An unknown fee still permits the gross result, but the corresponding after-fee amount remains unavailable. The model does not invent a broker commission from the selected currency or trade size.

When the purchase fee is known, acquisition cost is gross purchase value plus that fee. When the sale fee is known, net proceeds are gross sale value minus that fee. Net PnL and total fees become available only when both fee amounts are known. In the hypothetical example above, entry fee 2 and exit fee 3 give acquisition cost 202, net proceeds 237, total fees 5 and net PnL 35. Negative net proceeds can be a valid outcome when your stated sale fee exceeds proceeds; the model does not clamp this loss to zero.

4. Check the stated percentage denominator

The named net-PnL-to-acquisition-cost percentage divides net PnL by gross purchase value plus the purchase fee, then multiplies by 100. The exit fee is already deducted in the numerator. Both fees must be known before this ratio is available. Read its denominator literally: it is the model’s acquisition-cost comparison, not a claim to a universal ROI convention, margin return or annualized performance. It has no holding-period input and does not include dividends, coupons, tax, financing or later cash flows.

A loss larger than 100 percent can arise from explicitly supplied fees, even though the unleveraged sale price itself cannot be negative. The working division uses declared finite 384-significant-digit HALF_EVEN precision for recurring decimals. Cash products and differences use exact arithmetic. This distinction keeps the numerator separate from rounding of the percentage. Do not compare the resulting percentage directly with the annual return of a fund or a portfolio XIRR calculated over dated external cash flows.

1. Graham: match the earnings and book-value basis

Choose the explicitly named historical Graham-number variant. Supply your positive three-year average earnings per common share and the latest reported common-share book equity per share. Both must refer to the same comparable share basis and currency. Annual earnings from a single year are not the three-year average. The application does not fetch issuer statements, choose a sector or certify that your accounting inputs are comparable. Select suitability yourself; positive numbers alone do not establish it. Unknown or unsuitable comparability produces no screening value.

The displayed formula is the square root of 22.5 multiplied by average EPS and book value per share. The historical combined P/E and P/B limit explains the multiplier; it is not a current market benchmark. For an explicitly hypothetical EPS of 2 and book value of 20, the radicand is 900 and the screening number is 30 in the chosen currency. This arithmetic does not check the other defensive-investor criteria or establish a company’s intrinsic value.

2. PEG: keep current earnings separate from expected growth

Use current base-year EPS, the explicit growth-screen variant, your chosen horizon from one to ten years and your comparability assertion. Enter three independent growth percentages and three independent positive PEG targets for Bear, Base and Bull. For ten percent growth, enter 10, not 0.1. Growth must be positive and at most 100 percent. The scenario labels establish neither probability nor a ranking: you choose all six assumptions, and a PEG target does not default to 1 or come from your price.

The selected PEG multiplied by growth in percentage units gives model P/E; current EPS multiplied by that P/E gives the screening value. Hypothetical EPS 2, growth 5 percent and selected PEG 0.8 yield P/E 4 and value 8. The horizon describes the expected-growth assumption and does not multiply the price or compound EPS again. Supplying forward EPS would mix a different time basis into this model. Risk, payout policy and the quality of the growth assumption can affect comparability; the application does not verify those issuer facts.

3. Read your optional price comparison carefully

Your comparison price is a separate, optional positive input in the same currency. Leaving it unknown does not invalidate a supported standalone screening calculation. If you enable the field, supply a positive amount; a blank price is not zero. The difference relative to your price divides the model-minus-price gap by your price. The price gap relative to the model divides it by the model value. These denominators are different. A hypothetical model 30 and own price 15 therefore give 100 percent and 50 percent respectively; neither percentage is a realized investment return.

4. Explore explicit sensitivity instead of a hidden default range

First calculate a supported parent model. Then supply all three row assumptions and all three column assumptions yourself. Graham varies average EPS against book value. PEG varies growth percentages against selected PEG targets while holding the accepted current EPS, horizon, currency and suitability fixed. The nine cells are combinations of your choices, not nine provider forecasts. Values keep your order; repeated assumptions are allowed and remain repeated. No automatic plus-or-minus range, sorting or inferred middle scenario replaces your inputs.

Calculate sensitivity is an explicit action. Editing an axis hides the previous matrix until you calculate again. Editing or recalculating the parent resets its separate sensitivity panel. Invalid axes refuse the request. A cell exceeding the numerical budget is explicitly unavailable, with no guessed zero. Read row and column units with each value; changing two assumptions together does not identify which one caused the whole difference. The matrix creates no trades, account entries, probabilities, expected return or consensus among valuation methods.

5. Keep the result’s limits visible

Long displayed decimals marked ≈ are rounded presentation; exact working strings and declared numerical precision remain in detail. Inputs are your assumptions, with no invented provider provenance, issuer approval or market timestamp. The source links on the calculators explain the historical Graham relationship and the selected PEG approach; they do not certify your company inputs. Missing methods are not silently averaged into a six-model valuation. Use the breakdown to understand your model, and revisit weak assumptions rather than interpreting a single screening number as a buy instruction.

A time labelled Calculated on this device records your current explicit calculation. It is separate from the report periods behind your assumptions, a source publication date and a saved original’s creation time. No issuer timestamp is inferred from that local clock.

1. Choose how annual contributions enter the plan

Supply initial capital, your first annual contribution and a horizon in relative model years. Year one uses the contribution directly; its separate annual growth begins in year two. All money uses one chosen currency. These are your assumptions, rather than a historical plan, a broker offer or a calendar forecast.

A start-of-year contribution joins the balance before that year’s return and fee. An end-of-year contribution joins it after return and fee; it starts earning in the next year. The timing applies to the whole annual contribution, so this model does not silently turn it into monthly payments.

2. Keep the return assumption before the asset fee

Enter three independent nominal annual total returns for Bear, Base and Bull. Each rate is before the separately chosen annual asset fee and already includes whatever distributions your return assumption intends. Do not add dividends again or supply a return already reduced by the same fee. Negative returns are allowed above minus 100 percent; labels imply no ranking or probability.

The annual asset fee applies to the balance after that year’s return, before any end-of-year contribution. This is an explicit scenario convention, rather than a broker’s complete tariff. The prefilled amounts are editable model assumptions; mark a value unknown when you cannot establish it for your scenario. Supply a known zero if your assumption is no fee; an unknown fee leaves future net balances and totals unavailable. Deposits, signed modeled returns and fees remain distinct in the annual breakdown.

3. Compare nominal value with start-year purchasing power

Your inflation assumption builds an index from the start of the model. Divide each nominal closing balance by that year’s index to obtain modeled purchasing power relative to the starting price level. Zero inflation makes both values equal; negative inflation is a separate deflation assumption. This is not measured real wealth or a realized gain. Do not subtract nominal deposits from an inflation-adjusted balance and call the difference profit.

1. Supply the first model year directly

Enter your initial quantity and cash, first-year price per share and first-year gross annual dividend per share. These are personal assumptions in one currency, rather than issuer observations. Year one uses them directly. Dividend and price growth are separate annual percentages for each independent Bear, Base and Bull scenario; they start changing the supplied values in year two.

This narrow model has one assumed payment and reinvestment at each model-year end. It has no issuer payment calendar, regular contributions, inflation adjustment or split history. Growth may be negative but must remain above minus 100 percent. Scenario labels imply neither a ranking nor a probability. A dividend forecast is not a guaranteed distribution.

2. Inspect income and purchases separately

Opening shares earn the year’s gross dividend. Your supplied withholding and annual payment fee are deducted before reinvestment. The annual fee applies even in a year with no gross dividend. Cash carried from earlier years is available once. Shares newly bought at year end first earn a dividend in the next model year. Price growth changes purchase cost and valuation; it does not independently change the dividend per share.

Choose whole shares or fractions down to 18 decimal places. Purchases round down to the affordable quantity. The reinvestment fee applies only if a positive quantity is actually bought; otherwise the available cash carries forward. Closing wealth adds closing shares valued at that year’s assumed price and the remaining cash. Reinvested dividends are not added to wealth a second time.

3. Read incomplete and stopped scenarios honestly

The prefilled amounts are editable model assumptions; mark a value unknown when you cannot establish it for your scenario. Explicitly supply a known zero where your assumption is no deduction. Any missing deduction leaves the net projection and full-horizon totals unavailable, while first gross income can still be shown. If a payment would overdraw cash, the failed attempt is reported separately and does not enter completed rows or totals; the latest completed position retains its own year’s price. If the first attempt fails, only the starting position valued at the supplied first-year price remains.

1. Choose a consistent operating basis

Explicitly choose a nonfinancial operating business and a trailing or forward basis. Enter positive annual consolidated EBITDA and three independently selected multiples. EBITDA multiplied by each multiple gives operating enterprise value. No peer group, industry average, issuer statement or current multiple is supplied.

Use the same currency and accounting scope throughout. This narrow model rejects missing, zero or negative EBITDA and does not value banks or financial businesses. The declared basis is your assumption, rather than verified company classification or a reporting date.

2. Enter bridge items once

Operating EV excludes cash and nonoperating assets. To reach common equity, subtract debt, preferred claims, noncontrolling claims and additional nonoperating claims; add cash and nonoperating assets excluding cash. Additional claims must exclude operating payables already represented in EV and anything already included in another bridge field.

The prefilled amounts are editable model assumptions; mark a value unknown when you cannot establish it for your scenario. Select known only when you supply its amount, including an explicit 0 if your assumption is none. Net debt is debt minus cash and can be negative. All six amounts must be known before common equity is complete. This bridge is for operating EV; it is not added to DDM or FCFE again.

3. Distinguish an available value from a diagnostic

Positive complete common equity needs a supplied positive diluted common-share count before a model value per share is available. An unknown count leaves that value unavailable. Zero or negative equity remains visible as a diagnostic; it creates no positive share value, MOS threshold or price ratios. The supplied dilution is an approximation, without separate option valuation.

1. Enter the next annual dividend directly

Supply the positive gross annual dividend per matching share expected one year from now. This is DPS1, not the last payment or a quarterly amount. The model uses this input directly and does not multiply it by growth again. Annual dividends are modeled at year ends, with the first at year one; this is not an issuer payment calendar. All amounts share one chosen currency.

This personal dividend assumption does not establish an announced issuer payment. No dividend, tax deduction, buyback, earnings or free cash flow is supplied automatically. This narrow model values dividends, without substituting EPS or FCFE.

2. Keep discount and perpetual growth separate

For each independent Bear, Base and Bull scenario, enter your annual cost of equity and constant perpetual dividend growth. Value is next annual dividend divided by the difference between these rates after converting percentages to fractions. The discount rate must be strictly higher than growth; otherwise no model value is produced.

This is a stable perpetual-growth model. A valid rate difference alone does not establish sustainable growth, suitable distributions or a probability of success. The scenario names impose neither a ranking nor probabilities. Debt is not subtracted again from this equity model.

3. Compare assumptions and inspect the exact result

Your margin-of-safety percentage creates a separate threshold below each model value. An optional personal price supports two different comparisons: value relative to price and price discount relative to value. Without that price, those comparisons remain unavailable. The price never determines dividend, growth or discount assumptions.

1. Match the denominator to the multiple

Choose P/E for earnings per share, P/B for common-share book equity per share or P/S for revenue per share. Supply a positive amount in one currency on a consistent diluted share basis. Explicitly choose a past reporting period or forward assumptions for earnings and revenue; book equity refers to a past balance-sheet date or an assumed future date. These personal inputs do not establish issuer reporting dates or verified fundamentals.

The model multiplies equity per-share inputs. Enterprise-value multiples such as EV/EBITDA need a different denominator and equity bridge; they cannot be substituted into these fields. Zero and negative denominators do not produce a price here.

2. Supply three independent multiples

Enter a positive selected multiple for Bear, Base and Bull yourself. Each scenario value is the same supplied denominator multiplied by its own selected multiple. The names do not enforce an order or determine probabilities. No industry average, peer history or default fair multiple is supplied.

An optional supplied price is used only to compare the calculated scenario. It never selects the multiple. Without that price, price-relative percentages remain unavailable. Your margin-of-safety percentage reduces each scenario value to a separate model threshold; it is not a trading instruction.

3. Inspect the exact assumptions

Editing any input conceals the previous result until you explicitly calculate again. Open a headline to read its denominator, selected multiple, value, threshold and optional comparison. Exact detail preserves decimal strings, the method and actual calculation time.

1. Set the scope of this cost model

Enter one positive hypothetical trade amount, the portion used as a currency-fee base and an explicit number of monthly fixed payments from 0 to 120. All amounts share one chosen currency. The model does not fetch an exchange rate or convert a fee quoted in another currency.

A and B are personal assumptions, not current broker tariffs. Only one trade and the stated conversion portion are included; monthly fixed payments do not repeat the trade. Custody percentages, inactivity conditions, taxes and actual broker billing calendars are outside this narrow model.

2. Distinguish unknown, absent and known zero bounds

The unbounded trade fee is the supplied fixed fee plus the trade amount multiplied by the commission percentage. Then a known minimum is applied, followed by a known maximum. Explicitly select no minimum or no maximum when you want to omit that stage. A known minimum above a known maximum is a conflicting input.

An unknown rate, fixed fee or bound leaves the trade fee unknown. It is not a zero fee or an unlimited ceiling. A known zero must be supplied explicitly. A separately known raw fee can remain visible even when an unknown bound prevents the final trade fee.

3. Compare only complete totals

The currency component uses your stated portion and fee percentage. A portion of zero means no conversion in this model, rather than a verified zero tariff. Similarly, zero monthly payments means that component is not applicable; it does not establish a free account. Positive operations with an unknown fee remain unknown.

1. Choose a consistent cash-flow basis

This calculator uses positive free cash flow to equity per share, after financing. Supply an already consistent diluted per-share amount, one currency, the annual horizon and all three scenarios. It does not derive that amount from company statements or replace it with earnings, dividends or cash flow to the firm.

Discounting uses your cost of equity, and debt is not subtracted again. Constant dilution and growth within each stage are assumptions. The narrow model rejects a zero or negative starting cash flow; it does not claim suitability for every company or sector.

2. Read all three scenarios and their components

Bear, Base and Bull are names for your inputs, not guaranteed rankings or probabilities. For each scenario enter forecast growth, the discount rate and terminal growth. Annual cash flows are paid at year end. The model combines their present values with the discounted terminal value; the terminal growth must be strictly below the discount rate. This mathematical condition does not verify the economic sustainability of long-term growth.

Use the annual rows and separate terminal component to inspect what drives the number. A personal comparison price is optional; leaving it unknown does not invent a quote. The chosen margin-of-safety percentage changes a model threshold only. Neither that threshold nor the difference from your price is a verified value or an instruction to trade.

3. Explore sensitivity without filling missing assumptions

Optionally enter three increasing discount rates and three increasing terminal-growth rates for a sensitivity matrix using the Base forecast. Each combination is evaluated independently. An invalid rate relationship or numerical budget appears as unavailable, rather than a substitute price. Omit the matrix when you have not supplied its axes.

1. Start with your own first-year spending

Open the capital-withdrawal companion from FIRE & Coast FIRE. Enter initial capital, annual spending for the first model year, your annual inflation, effective annual return and the number of years. Use one currency throughout. The first spending amount is used as entered; inflation changes it only from the second model year.

These are constant personal assumptions. This tool does not fetch a market return or determine lifespan, taxes, fees, a safe rate or a probability of retirement success. The years are relative model periods, rather than an actual monthly spending calendar.

2. Choose the order of spending and growth

Start-of-year withdrawal first pays the supplied spending from opening capital, then applies the return to the remainder. End-of-year withdrawal applies the return to opening capital first, then pays the spending. These two orders can produce different balances with the same annual assumptions.

Each complete table row contains the opening capital, requested annual spending, model return, available capital at withdrawal, fully paid spending and closing capital. Editing an assumption conceals the previous result until you explicitly calculate again. This companion uses a distinct method from the accumulation and Coast target model.

3. Distinguish exact zero from an unfunded year

If available capital exactly equals spending, the full year is paid and the closing balance is exactly zero. A horizon ending there is covered in the model. Asking for a longer horizon can then produce a shortfall in a later year; that does not undo the earlier funded years.

1. Supply your own dates and amounts

Open Dividend reinvestment from Financial calculators. Enter initial quantity and cash, one assumed purchase price, dividend per unit per payment, your first assumed payment date, frequency and years. Use the same currency for all amounts. Neither price, payment nor date is fetched from an issuer or broker.

The price and dividend stay constant throughout this scenario. Payment dates step from your original month and day. A day beyond a month’s end is clamped for that month only, so a January 31 monthly anchor can return to March 31. Reinvestment is assumed on those payment dates, without a real trading or settlement calendar.

2. Distinguish unknown deductions and purchase rules

The prefilled amounts are editable model assumptions; mark a value unknown when you cannot establish it for your scenario. An unknown assumption leaves the first gross payment available but blocks the reinvestment projection. Enter each known percentage or amount explicitly, including 0. The payment fee applies to each model payment; the purchase fee applies only when at least some units are bought.

Choose whole units or fractional units. The calculator rounds the quantity purchased downward and carries remaining cash to the next payment. Fractional purchases use at most 18 decimal places. This is a model choice and does not establish whether a particular broker supports fractional units.

3. Read completed payments separately from a stopped scenario

The annual table groups your chosen number of payments into a scenario year and shows first and last assumed dates, quantities, gross dividends, deductions, each fee, purchase spending and remaining cash. Value at the assumed price is a derived model amount. It is not a live market valuation.

1. Enter your own consistent assumptions

Open FIRE & Coast FIRE in Financial calculators. Enter current capital, annual spending at today’s prices and the years to your target date. Choose one currency for all amounts. This choice labels the amounts and does not convert them. Enter 0 explicitly for zero capital, spending or inflation.

Supply annual inflation, an effective annual accumulation return and your assumed annual withdrawal fraction yourself. Effective return describes one annual growth factor. These fields are not populated from market history or a suggested safe rate. Valid negative growth or inflation remains negative.

2. Read the target and annual contributions

The target capital divides spending at the future price level by your withdrawal fraction. Coast capital discounts that target through your assumed annual growth to today. The existing capital projection shows growth without additions. The remaining gap determines equal contributions added at each year end.

Calculate scenario produces one result for the totals and annual table. The required annual contribution rounds upward to cents, and that same amount drives the table. Editing an assumption hides the old result until the next explicit calculation. Expand exact detail for the original decimal values, method and actual calculation time.

3. Keep a scenario separate from a retirement forecast

This model covers the years before the target date. It does not simulate retirement withdrawals, longevity, changing returns, sequence of returns, tax or fees. It supplies neither a probability of success nor a safe withdrawal rate. Changing your assumptions changes the arithmetic rather than verifying a forecast.

1. Enter a consistent payment assumption

Open Dividend income from Financial calculators. Enter the quantity, including fractional units if relevant, the dividend per unit for one payment and the number of equal payments per year. Use the same currency for the dividend and annual fees. Changing the currency labels the amounts; it does not convert them.

These are manual assumptions for one year. The calculator does not fetch announced company dividends or your actual portfolio, and it does not infer a share price, yield, future growth or reinvestment.

2. Keep unknown deductions unknown

The prefilled amounts are editable model assumptions; mark a value unknown when you cannot establish it for your scenario. If you know your scenario assumptions, choose the explicit option and enter the percentage and amount. Enter 0 when you explicitly assume none. A blank or unknown field does not silently become zero. The model does not determine legal tax rates or tax obligations.

Gross income multiplies quantity, payment per unit and frequency. Net income is shown only when both deductions are known. Fees can exceed the remaining income, so a negative net result stays negative. The monthly equivalent is annual income divided by 12, rather than a schedule of actual payments.

3. Review the exact scenario

Calculate income produces one result used by the total, breakdown and monthly equivalents. Editing an input clears that result until you calculate again. Expand the exact detail for original decimal strings, method version and the actual calculation time. Formatted totals can display fewer decimal places.

1. Start with the portfolio and account

After signing in, create your portfolio and choose its accounting base currency. Add the real accounts you want to record. Choosing CZK, EUR or USD does not convert a transaction, verify an exchange listing or activate a market-price source.

Organize portfolios beside the selector previews Pin/Unpin and keyboard Up/Down within pinned or other portfolios. Pinning moves an item to that group’s end; changing order never selects another portfolio. Reset uses the current complete owned inventory without creating objects. Save is explicit and owner-versioned; Cancel returns to the admitted saved order. A changed inventory or revision preserves your draft until an explicit new review. An uncertain request retains its exact identity; stopping retry at a fresh later revision does not prove or undo its old outcome. Organization changes no accounts, currencies, names, economic events or market values; JSON copy/restore and duplication are not provided by these controls.

Copy / restore creates a new portfolio; it never overwrites the selected ledger. Download the separate restorable JSON or paste that whole archive, enter a new name, and review its accounts and ordered events before explicit Create. Back and Cancel create nothing. The whole-copy limit is 50 accounts, 500 events and 1 MiB; an oversized archive is refused without truncation. Original amounts and effective times remain readable, while accepted event storage uses the existing kernel normalization. Historical clocks and ledger/event hash descriptors describe the source, not the new write time or equal destination hashes. The new archive checksum binds the portable material only. No market valuations, external quotes, access proof, retry identities or CSV jobs are included. An uncertain Create retains the exact original command for retry; it is not a confirmed creation.

Instrument identifiers entered in your ledger belong to your own records. Keep an account and currency consistent with the original transaction. A ticker alone does not establish a particular exchange, fund share class or trading pair.

2. Record or preview before confirming

In Holdings, Record buy or Record sell opens a proposal for that exact own instrument and accounting currency. Opening it preserves unfinished inputs. Prepare a new blank draft explicitly replaces them with this context and otherwise blank financial fields. A single current open-lot account can be prefilled; with multiple accounts, select one explicitly. Enter the actual quantity, unit price, effective time and any fee yourself, then use the existing Review and Confirm. No quote or full-position sale is inferred. An expired or changed ledger proposal must be reopened from a currently verified position; an existing review or pending write cannot be replaced.

For a manual transaction, enter the actual effective time and exact amounts, then choose Review transaction. Check the selected account, currencies, fees and both the local-zone and UTC time. Back to edit retains the inputs without writing; Confirm transaction entry is the separate action that posts the immutable event.

Record an economic event explicitly, including its amounts, currency, fees and effective time. Accepted events form an immutable journal. Corrections use a reversal rather than editing a historical record. Backdated changes require controlled replay and can be refused by the current pilot.

The import screen provides a universal CSV header. Prepare your own records in that format and review the normalized events, duplicate decisions and warnings before confirming. Your own account ID connects each row to the intended account. Broker-specific formats need their own verified samples; the universal import is not a promise of automatic compatibility with every broker.

Multiple CSV accepts two to six original files in that universal format. One combined preview checks duplicates across the files and the existing journal, while keeping each file and logical record identifiable. Review the complete bundle before confirming it together. History completeness is your declared assessment; it does not supply missing prices, exchange rates or acquisition costs.

Saved history lists imports for the selected portfolio. Open an original receipt to distinguish newly added events from duplicates and ignored records, and read its original confirmation time. A preview or expired import has not posted those events. Older single-file imports without an original membership receipt remain visible with an explicit limitation; their added events are not reconstructed from a current preview.

Undo is available only when the receipt proves the original new events and those events are still the last active operations in the journal. Later active operations, missing evidence and previously reversed events can block it. Review the exact events and proposed opposite journal entries first. Back writes nothing; Confirm undo appends linked reversals while retaining the originals and their receipt. Duplicates that already existed before the import are kept. This changes your private records only; it sends no instruction to a broker. If the outcome after confirmation is uncertain, keep the exact request in the open page and explicitly verify or retry that original confirmation; a new request does not replace it.

If a submitted bundle has an uncertain outcome, keep its original request for an explicit retry. Discard requires its own review and resolves that same bundle: if confirmation already succeeded, the accepted receipt remains and the journal entries are not undone. Start a new preview only after the original outcome is resolved.

3. Trace the numbers back to the journal

Open account detail brings its recorded cash, current open lots, realized sales, transfers in both directions and complete related records together. A transferred lot belongs to its current account while its original acquisition keeps its original account. Open the original event from a lot, or inspect a sale’s FIFO; that inspector also labels its whole-portfolio instrument scope. Unknown cost basis remains unknown. The account ledger includes incoming transfers and their linked reversals.

Account JSON and CSV include every selected-account section and display page, with separately labelled cross-account acquisition provenance when needed. Each download verifies the same held economic cutoff and complete source again. The ledger hash describes the full original portfolio, not the subset file. These reports do not allocate whole-portfolio market value or investment return to an account and are not restorable backups.

Recorded cash, quantity and realized profit are different from complete market value and investment return. Missing verified prices or dated FX conversions can prevent a valuation in the base currency. A recorded zero is not a substitute for an unavailable value.

Cash also separates recorded external deposits and withdrawals from your current balances. Deposits and withdrawals are shown as positive magnitudes; net cash capital is deposits minus withdrawals in each original currency. Account and UTC-month filters narrow the selected original records only after Apply. Transfers between your own accounts, trades, fees, FX and dividends are not external cash capital. External securities receipts are counted separately without inventing their cash value. Never add totals from different currencies or read net capital as profit or return.

The selected economic cutoff includes only events and reversals effective through that instant. A later reversal therefore does not remove a deposit from an earlier cutoff. This is a view of today’s saved ledger at that economic time, not proof of what was known then. The applied cutoff stays fixed during verification; draft edits require Apply. Open an original event for its recorded details. JSON and CSV export all matching rows and the same exact totals, cutoff and source identity after a fresh check. A changed source requires a new explicit verification rather than silently substituting a newer ledger. These private reports are not restorable backups or tax conclusions.

Dividends shows active recorded dividend events through the explicitly applied economic cutoff. Gross, recorded withholding and net are exact separate totals for each original currency; net is gross minus withholding only. Monthly cash groups use the recorded effective time in UTC, matching the existing ledger convention. Original account and instrument identities, the entitlement time and optional settlement time remain separately readable. Entitlement is your recorded claim, not a verified ex-date, and settlement does not silently change the cash convention. Future reversals do not erase earlier cash before their effective time. No future payments, yield, currency conversion or tax conclusion is inferred; fees and account cash balances remain separate.

Choose the current or a fixed UTC instant and select Apply. The server creates one whole report from the exact admitted recorded ledger at that economic cutoff. Draft edits leave the applied report unchanged. Apply alone retains only a temporary result; it does not save a permanent Snapshot. Currency totals, UTC months and original dividend rows all belong to that same result. An economic cutoff does not reconstruct information known then.

To retain this original, choose Save Snapshot, enter its name and inspect the separate save review. Check the exact portfolio, source ledger identity, cutoff and table presentation, then explicitly Confirm save. Back or closing the review saves nothing. The Snapshot retains the whole original report and own inputs; later ledger entries and account or portfolio renames do not overwrite its recorded amounts or original names.

In My analyses choose Open original analysis to return to the exact saved dividend report. Its dedicated /portfolio/reports/dividends?acceptance=… destination opens that original, without replacing it with a current calculation. Verify original overview checks access to the same retained result. A changed or unavailable account or private source can prevent reading; a saved identity is not an access grant.

Whole report JSON and CSV use a fresh authorized server download of the same original result. They include every currency total, month, dividend and original source event, including records outside the economic cutoff and beyond visible pages. Original event details preserve economic, entitlement and optional settlement clocks separately. Write retry keys and access proofs are omitted; displayed source hashes describe the original private source, not checksums of the sanitized file. The export is not a restorable backup, a forecast or a tax return.

To continue with a new calculation, choose Prepare cutoff adoption into a new draft and verify the current original portfolio. Inspect the proposed cutoff and explicitly consent before Adopt cutoff without calculation. An existing draft additionally requires explicit replacement consent. Only the time is adopted: no old prices, amounts or account selections are copied, and nothing runs automatically. The original report stays separate; use Apply when you deliberately want a new report from the currently admitted ledger.

Costs separates recorded fee charges, reversal credits and signed net costs in each original currency. It includes standalone fees and the recorded fee legs of buys, sells and currency conversions. Account, currency and UTC-month controls remain drafts until Apply; the chosen economic cutoff stays fixed across verification. Monthly and account summaries and whole-selection exports use the same original accounting lines. Open a row to inspect the economic operation that created it; a reversal retains its original link. Missing fee inputs do not prove a zero broker charge. Dividend withholding stays in Dividends, and these costs must not be deducted again from an already fee-adjusted FIFO result. This is recorded accounting, not a tax liability or total investment-cost estimate.

Download ledger JSON checks the current portfolio again and exports its economic events and original audit references. It is a private audit file, not an importable backup, a market valuation or a tax return. Keep the file with the same care as the records it contains.

The Ledger’s applied account, instrument, type and inclusive UTC date selection can become a separate private selection report. First Apply the filters; draft edits are not included. Review names the actual source portfolio, shows the whole and matching record counts and makes all original matches available in fifty-row display pages. Download JSON verifies the current account and the same complete source again, then includes every match, not just the displayed page. Original recorded sequence is reversed exactly as in the Ledger; future and reversal records follow its existing selection rules. A changed source requires an explicit new review. Source hashes describe the whole ledger, not a subset checksum. This selection report is not a restorable backup, market valuation or tax conclusion; Back and Cancel write nothing.

4. Compare two recorded portfolio states

Open Changes over time from the portfolio overview or its section navigation. Enter two ordered UTC instants and explicitly Apply. Both states are calculated from the same complete currently recorded ledger, using the original portfolio projection rules. Editing either instant leaves the applied report unchanged until a new confirmation. A changed ledger requires a new application; it is not silently substituted beneath the old report.

Read quantities before, after and their exact signed difference for each instrument and accounting currency, with the account detail needed to explain internal transfers. Cash stays separated by account and original currency. A move between your own accounts can change the account rows without changing the total instrument quantity. A split can change quantity without creating investment wealth. Unknown acquisition basis stays unknown.

Each endpoint includes events effective at or before that instant. The interval event list therefore includes events strictly after the first endpoint and at or before the second. Inspect economic time separately from the original recorded time and follow reversal links to their original events. A later-effective reversal is absent from the earlier state. Backdated events recorded later can affect both retrospective states; this is not a reconstruction of what you knew at the earlier date.

The full JSON and CSV preserve the same applied report, both cutoffs, exact source identity and original records beyond the visible page or search. Export rechecks current access and that exact ledger. The report changes no transaction and needs no market-price source. Cash difference is not net contributions, profit or investment return; native currencies are not summed into a fictitious total. This is not a historical market-value chart, TWR, XIRR or tax result.

1. Describe your scenario

Choose a currency, initial capital, nominal annual rate and number of years. The compound-interest calculator adds the contribution at the end of each selected compounding period: yearly, quarterly or monthly. A contribution of 1,000 per year is a different plan from 1,000 per month.

The rate is your constant assumption. It is not a current bank offer or a predicted investment return. The currency labels the amounts; changing it does not convert them with an exchange rate.

2. Read the result in context

The projection chart separates the model balance from your accumulated contributions. Move its year selector with the keyboard to inspect the same annual rows as the table. Straight connections show the shape between yearly points; they do not add monthly values or a market backtest.

The final balance separates your contributed capital from interest. Contributed capital includes the initial amount. The effective annual rate shows the effect of the chosen compounding frequency on the nominal rate. The year-end table follows the same assumptions.

Target savings uses monthly compounding and end-of-month contributions. The required monthly contribution rounds upward to currency cents; that same rounded payment is used in the displayed balance and annual rows. If the initial capital already reaches the target under the assumed rate, the required contribution is zero.

Fees, taxes, inflation and changing rates are outside this model. Compare alternative rates by entering your own sensitivity assumptions. The output describes the calculation, not a guaranteed outcome.

3. Preserve the assumptions

Save to file downloads a local scenario containing its method version and inputs. Load file validates those inputs and lets you calculate them again. The file does not upload your finances to a public board or save a scenario to an account.

Four clocks, four meanings

The observation period is the day or month described by a value. Source publication is when the provider released it, if that time is documented. First observation is when Datisto actually captured it. Query execution is when your analysis was calculated. A recent query does not make an old source observation recent.

Unknown publication times stay unknown. The precise recorded timestamp remains available beside the formatted date. A chart can use a coarser axis while the table and Passport preserve the original recorded precision.

Latest history and earlier knowledge

Latest uses the revisions available today. Known to Datisto then uses the platform’s actual capture history up to your cutoff. Publicly known then requires documented source publication times. These modes can produce different selections and are not interchangeable.

An unavailable value is not zero. A missing month is not silently filled. Source-reported estimates and revisions keep their own quality labels. Compare a retained analysis with a new calculation when you want to inspect how its source history changed.

Availability is separate from permission

The Sources page distinguishes display, derived analysis, retention and download permissions. A readable value does not automatically permit every export or public reuse. Current source rights are checked again when an analysis is opened.

Current pilot coverage includes selected reference exchange rates, inflation series and model configuration. Reference rates describe the provider’s published reference, not an executable trading quote. Model configuration is not a hosted-service price offer.

Use integrations and setup without confusing their states

On Sources, switch between Datasets and coverage and Integrations and setup. The integration view describes 25 static implementation profiles, with search, category, access and usage filters, official documentation and links to their results. A keyless/optional/server-key-required label describes that profile’s requirements; it does not mean connected, licensed or currently available.

What is currently recorded shows only exact linked datasets, recorded permission status and observed stored-series/period counts. Unknown mapping, a failed read and no linked dataset are distinct states. Coverage does not prove current provider health or complete history. Open dataset permissions, series and archive returns to the existing directory and its original source rules; result reads and exports still make their own access checks.

For OpenRouter · AI, explicitly choose Check assistant configuration. This reads the existing service capability only: disabled, missing key or configured. A failed or unchecked state stays unknown; configured does not establish a successful model call or answer quality. The check neither reads a key nor sends a model question. Other profiles have no inferred connected status and are not probed by fetching their data.

My preparation stores optional checklist notes only in this browser. Checking a box does not register a service, configure the server or grant data rights. Operator examples contain variable names and placeholders; no secret is entered here and examples do not execute themselves. Open the relevant main result to inspect actual data and source clocks. Specialist previews remain labelled where no equivalent main tool exists.

1. Separate your reasoning from the financial ledger

Choose Decision, Review, Note or Lesson and mark the entry Open, Reviewed or Archived. These describe your own record; Reviewed is not editorial approval and Archived does not delete it. Write a title and plain text. A journal entry does not create a trade, deposit or financial-ledger event.

Optional entry-price bounds, target price and currency, invalidation conditions, desired portfolio weight, horizon date and confidence are your assumptions. They are not observed market prices, actual portfolio weights, recommendations or calibrated probabilities. A blank optional field is unspecified, rather than zero. Use a decimal point for numbers; choose a currency when entering a price.

2. Attach only your verified current context

You can leave an entry unbound or choose a portfolio belonging to your current account. An instrument must be an exact identifier already recorded in that portfolio’s financial ledger. A link from the portfolio offers context; explicitly apply the verified context instead of assuming that opening the link has filled or saved anything.

A recorded instrument identifier is not a verified ticker or a data licence. Reading a current entry or an older version rechecks its own context. An unavailable portfolio or instrument is not replaced with a plausible label or another account’s data. Your own event date and time are distinct from the application’s original storage clocks.

3. Review the exact words before saving a revision

Review draft checks the complete fields and shows the exact content and expected revision. Back to edit preserves your words; it performs no save. Confirm private save is a separate action for the displayed account and session. Each accepted save creates a new immutable version of the same entry. The saved-version history keeps the earlier words and storage times.

If a newer revision exists, read it and explicitly review your retained words against the current revision. The application does not silently overwrite it. If the save outcome is unconfirmed, the original command remains separate from a draft; verify the same account and explicitly retry the original request. Editing words does not change that retained request.

4. Resume or discard a same-tab draft explicitly

Unfinished raw text can remain in this tab while you navigate within the application. The shared temporary memory holds at most eight drafts in total, with a 30-minute expiry from the latest remembered edit. Only your freshly verified account’s drafts are shown. This is not persistent storage: a reload, closing the tab or expiry can lose the draft. There is no transfer of drafts between accounts.

Use Resume to restore the original entry identity, raw words and base revision, including incomplete number or date text. Validation waits for Review; resuming does not save or silently rebase the draft to a newer revision. Discard removes only the exact own temporary draft, not a saved entry or an unconfirmed save command. Review, a running save or a retained pending command blocks conflicting draft actions. Context offered by a route does not replace resumed context automatically.

5. Know what your current JSON download contains

Download current own JSON rechecks the current account and exports the current saved contents available to that account. Unfinished drafts and the full version history are not that export. Keep the original file private if it contains your reasoning or assumptions. A download is not confirmation that an unfinished save succeeded.

The separate journal JSON import reviews current contents and lets you choose Skip, Copy as new or append to the current revision. It does not restore the complete historical audit trail. Reminders and goals have their own separate screens and explicit confirmations; opening a journal entry does not create them. For a selected saved goal, separately review and confirm one own saved annual plan in the same currency, or remove that association. This changes neither goal words nor manual state and measures no progress. An unavailable plan stays neutral without its private metadata; the goal remains readable. Temporary drafts still require Review and confirmed private Save to become saved notes. Public sharing and a tested full-history restoration backup are not provided by this workflow.

6. Record a separate review of a decision

Open a currently verified saved Decision and choose Review this decision. Enter your own new title, review date, conclusion, reasons and evidence, any changes, and your next action. An optional next review date cannot precede the review date. Your conclusion is your own assessment, not an automatically verified market outcome.

Preview new review shows the entire generated plain text. It cites the original entry ID, revision and title. This is a text reference, not a live association, a history of source values or a reminder. No prices, original assumptions or investment horizon are copied. The next review date remains text; a day-only review date does not invent a midnight event timestamp.

Adopt as a separate draft creates a new local entry identity with kind Review, status Open and base revision zero. Only currently admitted private portfolio and instrument references are carried over. Your original decision, its status and financial ledger remain unchanged. Other unfinished words and drafts are preserved; if temporary draft capacity cannot accommodate adoption, your review fields remain and adoption is refused. Continue through the journal’s usual Review draft and separate Confirm private save; adoption itself saves nothing to the account.

If current verification is temporarily unavailable or the decision changes, your form is retained but adoption is blocked. To use a newer source revision, explicitly adopt that currently verified revision as the source while retaining your own fields, then inspect the full preview again. The form exists only in this page’s memory. After adoption, the ordinary same-tab draft limit of eight drafts and 30-minute expiry applies; reloading or closing the tab is not durable saving.

1. Choose the owned portfolio and your terminal assumption

Open Portfolio, select your own portfolio and its Model section. The XIRR panel uses the current admitted ledger revision and complete external-flow table for that portfolio. It does not borrow another account’s records or accept a selected handful of convenient flows. Verify the current portfolio before calculating if its source or account cannot be checked.

Enter your nonnegative total terminal value, including cash, in the portfolio’s base currency. This is your valuation assumption, not a fetched market quote, recorded acquisition cost or verified portfolio value. An explicit zero is valid; a blank field is missing. The field accepts a decimal point, up to 24 integer digits and 18 fractional digits.

Supply an exact valid UTC instant with milliseconds, using YYYY-MM-DDTHH:mm:ss.sssZ, strictly after every recorded external flow. There is no default valuation date. The supported calendar is 2000–2100 and the maximum span is 36,500 days: 100 model years of 365 days, not a promise to accept every hundred-year calendar interval. Then explicitly choose Calculate own model.

2. Read cash-flow direction and numerical availability

Cash contributions have the investor’s negative direction; withdrawals have the positive direction. Your supplied terminal value is the final nonnegative valuation flow. Internal purchases, sales, dividends and fees are not added again as external contributions. Inspect the original flow table, base currency and ledger revision rather than confusing account cash movements with investment performance.

The current model supports base-currency cash flows only, at most 50 complete flows. Foreign-currency cash flows and securities receipts are unsupported, even when another view has a base amount. Missing historical FX or unsupported source widths also refuse calculation. Your terminal assumption cannot repair that history. The application never takes the first 50 rows, guesses a price or truncates money to fit. An unavailable result is never replaced with zero.

The solver uses finite 48-significant-digit numerical precision and an Actual/365 basis. Its disclosed search interval [-30,30] bounds log(1+r), not percentages. Only a unique conventional root becomes an available rate. Multiple or nonconventional roots remain unavailable; detected roots are lower-bound diagnostics. No root within the search interval does not rule out roots elsewhere, and an eight-second timeout does not prove that no root exists.

3. Keep the model separate from actual performance

Read the approximate displayed percentage alongside its original numerical rate and methodology. The raw root is the finite working output, not an infinitely exact real-number solution. It is a model conditioned on your terminal value, not certified market performance, TWR, benchmark comparison or tax output. Realized sale PnL measures a different quantity. Microsoft XIRR documentation explains irregular dated flows and a 365-day basis, but spreadsheet date truncation and its iterative solver differ; no Excel parity or external certification is claimed.

Editing either terminal input conceals the old model and requires another explicit calculation. Account, session, portfolio, visibility or source changes can also hide it or cancel ongoing work. A fresh authority check may readmit the same completed numerical bundle when its exact inputs and recorded flows are unchanged; that check does not silently recalculate a return.

Calculating creates no ledger event, source quote or account-saved XIRR original. The manual-input panel has no file import or export. Use the private investment journal to record your own explanation if needed; it does not attach or certify the computed rate automatically. Local inputs are not a backup and can be lost on reload or leaving their scope. This guide describes application behavior, not editorial approval or investment advice.

4. Use a complete original valuation as the terminal input

In Valuation, prepare the complete single-currency report in the portfolio accounting base currency. Its inline source-terminal XIRR panel can explicitly adopt that whole total, including cash, and its exact economic cutoff. It never substitutes the known portion or cost basis. Review the original source identity and rounding before calculating; a changed original, even with an equal total, requires a new explicit calculation. The separate manual Model section is unchanged.

This uses the exact reported total, including any already disclosed HALF_EVEN rounding to 18 decimal places. It does not round again to fit XIRR limits or move the cutoff to bypass a same-time flow. The complete flows come from the same recorded ledger and applied cutoff, including a custom earlier cutoff. Foreign-currency historical flows, securities receipts, unsupported widths, more than 50 flows and ambiguous numerical roots remain unavailable under the existing rules.

Read this as a model with chosen original source prices. Prices and CNB fixing have their own dates and need not be contemporaneous with the economic cutoff, so this does not establish verified actual portfolio return. The complete model JSON verifies the same ledger, current session and original source download access again, then includes the original valuation, all modeled external flows, terminal input, numerical result and methodology. It omits session and authority-envelope fields and is not an executable worker command or restorable backup. Loss of source access or a changed account hides the model and prevents download.

1. Equal-amount DCA and a generalized schedule

Investor.gov defines dollar-cost averaging as investing equal portions at regular intervals, buying more units when prices are lower and fewer when prices are higher. This supplies a definition only, not a claim that a strategy is superior. The application also accepts unequal contributions and irregular explicit dates: those inputs describe a generalized dated cash-purchase schedule rather than strict fixed-amount DCA.

Choose the model, currency and whole units or the fractional step 0.000000000000000001 explicitly. Enter initial cash, including an explicit zero when appropriate. Add one to sixty strictly ascending civil dates from 2000 through 2100. Every contribution and assumed unit price must be positive. Prices must describe the same comparable unit and currency; selecting a currency never converts them.

2. Read carried cash and conditional fees

At each date, opening cash plus the contribution becomes available cash. With a known fee, the model buys the largest quantity fitting that cash at the selected step. The whole-purchase fee applies only when a positive quantity is bought; a zero purchase pays no fee and preserves cash. Unspent cash is carried to the next supplied date. The next-step cash requirement shows why one further quantity step cannot fit.

A blank fee means unknown, not zero. Enter 0 for a known zero fee. Any unknown fee blocks the entire purchase projection: only the total supplied contributions and missing fee rows are shown, without partial quantities, balances or averages. Calculate explicitly after editing; edits, adding or removing rows and Clear conceal the old result.

3. Separate the two averages and the model limits

Weighted execution-price average is purchase spending divided by total units. Model acquisition-cost average adds applied purchase fees to spending before dividing by units. Both are unknown when no units were bought. They are distinct model ratios, not a tax cost basis or an investment return. Exact cash and quantity calculations retain carried fractional cash; recurring ratios use finite 384-significant-digit HALF_EVEN precision. A displayed ≈ marks presentation rounding only; exact working output and original input strings remain in method details.

1. Define the constant-window assumptions

Choose the Bitcoin difficulty-convention model and currency explicitly. Supply a positive hash rate in TH/s, active power in W, uptime from 0 to 100 percent and one to 366 days. Active time, hashes and energy use these unchanged assumptions over the entire horizon. The tool supplies no miner specification, network values, price, wallet, pool contract or hardware recommendation. A fiat currency label does not perform conversion.

Known network difficulty must be at least 1. This is Bitcoin mainnet difficulty, not a pool share-difficulty setting. Your gross reward per whole block includes the assumed subsidy and transaction fees. The percentage pool fee is a simplified deduction, not a description of a real pool payout policy.

2. Keep unknown assumptions distinct from zero

Blank difficulty, reward, pool fee, electricity tariff or assumed BTC price means unknown. Enter 0 for a known zero where allowed; difficulty cannot be zero. Every result lists its own missing prerequisites. An unknown tariff remains an unknown electricity cost even at zero energy. A known price of zero, reward of zero, uptime of zero or pool fee of 100 percent has its explicit model meaning but never fills another missing assumption.

Active hours are days times 24 times uptime/100; energy is active power in W divided by 1000 times active hours. A supplied tariff gives exact finite electricity cost. Offline, idle and cooling energy, fixed tariffs, hardware, maintenance and tax costs are excluded. The model difference subtracts only that electricity cost from expected after-pool fiat proceeds; it is not total profit, ROI or payback.

3. Read the approximation and original details

The chosen convention divides active hashes by difficulty times 4294967296 (2^32). This is an inferred conventional approximation, not exact target probability or reconstruction of current nBits. Bitcoin Core work uses 2^256/(target+1), while the difficulty-1 target is 65535 times 2^208. Public Bitcoin documentation supplies those concepts, not current network figures or certification of this simplified formula.

Expected blocks, BTC and their fiat values use finite 384-significant-digit HALF_EVEN working precision and are displayed with ≈. Expectations are not guaranteed blocks, rewards or payouts; the model supplies no variance or confidence interval. Exact time, hashes, energy and supplied electricity cost are labelled separately. All summaries and rows use one immutable result, whose original strings, precision and actual device calculation time remain readable in details.

1. Define the units and count each cost once

Select the own isolated linear variant and long or short explicitly. Enter a positive underlying quantity q and a positive entry price E per unit. Choose CZK, EUR or USD as the common quote currency. Quantity is not a contract count with an inferred multiplier. All collateral and reserve amounts must already use that currency; the label performs no conversion. The variable P is a hypothetical positive valuation price, not a fetched market observation, executable quote or exchange mark price.

Collateral C is nonnegative and is supplied BEFORE the separately entered fixed cost reserve R. That reserve has not already been deducted: the model subtracts it once. Supply your constant maintenance percentage m and, if known, closing-fee percentage f. They apply to q times P in this chosen model. Enter percentages as percentage units, not fractions; the calculation divides by 100. Each rate is below 100 and a known sum m+f must also be below 100.

Blank closing fee or reserve means unknown. An explicit 0 is a known zero assumption. Missing costs leave the complete entry buffer and boundary unavailable; they do not become zero even when another metric is zero. Independently known entry notional qE can still be shown. These are your supplied assumptions, not verified account balances, tariff rules or accrued funding charges.

2. Read the equation and its distinct states

For long, model equity is C−R+q(P−E); for short it is C−R+q(E−P). The model requirement is qP(m+f), with m and f now fractions. At entry, compare C−R with qE(m+f). A negative buffer means the requirement is already unmet at your entry assumption. An exact zero is an at-entry model boundary. Neither state establishes an actual margin call, forced close or permission to open.

With a positive entry buffer and known costs, equality gives long P=(qE+R−C)/(q(1−m−f)), or short P=(C+qE−R)/(q(1+m+f)). Read the numerator and denominator separately. A valid long numerator at or below zero produces no positive crossing in the chosen P>0 domain. Its boundary stays null, rather than a fabricated zero price. This result does not guarantee that a real position is safe.

Finite products, cash differences and rate fractions use bounded exact arithmetic. Only the final ratio uses 384 significant working digits with HALF_EVEN rounding and is labelled approximate. The raw boundary string is a finite working result, not infinite real-number precision. A numerical budget refusal returns no partial boundary. All displayed components belong to one immutable result with the original entered strings.

3. Recalculate deliberately and keep real venue rules separate

CME material on futures margins and Bybit material on margin modes and liquidation explain concepts and omitted scope; they do not certify these independently chosen equations. Actual products can involve initial-margin admission, maintenance tiers, mark-price rules, cross or inverse contracts, funding, collateral haircuts, insurance and liquidation procedures. This model supplies none of those rules or current values. It is authored application help for the internal pilot, not editorial approval, a venue calculation or a trading recommendation.

1. Compare one household consumption band

The selected Eurostat NRG_PC_204 series describe household electricity in band DC: annual consumption of 2500–4999 kWh. Each observation is a weighted national end-user price for that band, expressed per kilowatt-hour. It brings the three countries onto the same statistical scope. The band describes annual consumption, even though each price observation covers half a year. It is not a prediction of how much a particular household will consume.

Use this view to examine how the recorded national prices differ or evolve within that common band. A national average does not identify a supplier, contract or household. It also cannot establish the total cost of a particular bill without that household's actual consumption and contractual terms.

2. Read what is included in EUR/kWh

These series include all taxes, levies and VAT. The source methodology also incorporates subsidies and allowances. Changes can therefore reflect support measures and taxes alongside energy, supply and network costs. The displayed value is a period statistic, not today's supplier tariff, a wholesale bidding-zone price or an electricity-cost assumption already chosen for a mining model.

Eurostat converts national-currency prices into euros using the average exchange rate for the reported period. An EUR/kWh movement can consequently combine a domestic price change with an exchange-rate effect. Datisto displays the source euro value without adding its own currency conversion. Keep that convention in mind before interpreting a difference between countries or periods as a change in local contractual prices alone.

3. Separate the half-year period from the clocks

S1 covers January–June and S2 July–December. A point's period identifies the half-year represented; it does not mean that the price was available on the period's first day or stayed unchanged every day. Compare aligned semesters and read the retained coverage rather than treating the series as daily observations.

Distinguish four clocks: the source dataset update, the individual observation's publication, Datisto's capture and Datisto's input-record time. Individual publication is unknown here; a dataset update cannot supply it. Capture records when Datisto obtained the response, while the input record marks its ingestion. Historical statistics can be revised, so a later capture is not proof of what was publicly known at an earlier date.

4. Open the raw comparison and inspect a point

Open the three-country comparison to use the original EUR/kWh series with the same band and tax treatment. This dataset supports raw values only. The chart and observation table share one result; gaps remain gaps. A source flag p means provisional, not missing or zero. It stays attached to the observation and can later change with a source revision.

Select a value to open its Data Passport. Check its original amount, unit, period, quality and source evidence, then distinguish the recorded clocks from the current access check. Source details explain the dataset's scope and permissions. These steps help establish what a comparison contains; they do not turn a statistical average into a personal offer or certify a particular supplier's price.

1. Read the quantity before the rate

Every row states how many units of the foreign currency the Czech koruna amount refers to. Most currencies use one unit, but some use one hundred or one thousand. A large or small displayed number therefore does not rank currencies by strength. Read the currency code, original quantity and CZK unit together before comparing two rows.

For a concrete recorded example, the captured fixing for 14 October 2026 reports 13.730 CZK for 100 JPY. Datisto retains the original 13.730 string; the table may format it with four decimal places and a local decimal separator. It does not turn that stored observation into a rate for one yen. IDR uses 1000 units. This example describes that dated source record, not a current price or an offer to exchange money.

2. Choose an exact day without filling gaps

The proposed day comes from the common stored coverage of all thirty reviewed currencies. It is not inferred from your device date. The latest stored fixing loads automatically, as does a day you select. The table asks for that civil date only. A weekend or another date without an observation remains unavailable; it does not receive Friday’s rate or a zero.

The current source scope covers 2026. A fixing date identifies the day described by the observation. It does not prove the exact publication time, nor that Datisto already possessed the rate on that day. Opening the history shows the retained original observations over their actual coverage, rather than generating an observation for each calendar day.

3. Search, inspect and export one page

Search accepts a currency code or a Czech or English country or currency name. For example, “jen” finds the Japanese yen. The thirty currencies occupy four pages, with at most eight on an applied page. Changing the day, search or page hides the previous result. The new selection loads automatically, with search waiting briefly while you type. An older reply cannot appear under the new controls.

Open Value provenance beside a rate to inspect the original value, quantity, period, quality and source evidence. The page export contains the same applied result and its manifest. It includes that page only; exporting eight rows does not mean all thirty currencies were exported. Moving to another page automatically reads its own result, rather than one simultaneous snapshot of the whole sheet.

4. Separate the clocks and the use of the rate

The Passport distinguishes the fixing period, publication when known, Datisto’s actual source capture, input ingestion and result computation. Individual publication is unknown in this captured source. A later access check verifies whether you may read an existing result now; it does not change its original computation time or establish an earlier publication.

Source: Czech National Bank. The sheet provides reference observations, not bank buy/sell spreads, exchange fees, intraday trading prices or a promised execution rate. This workflow preserves raw source rates and does not perform cross-currency conversion. Open the source panel and original document when the exact definition matters, and keep another publisher’s rate separate unless the units, date and methodology have been reconciled.

1. Identify the denominator and adjustment

The harmonised unemployment rate is a percentage of the labour force aged 15–74. The source code TOTAL refers to that total age scope; it does not mean every resident. This is neither a percentage of the whole population nor a count of registered jobseekers. Keep that distinction when comparing a national headline with this series.

The three selected series are monthly, cover both sexes and use SA: seasonally adjusted, without calendar adjustment. Matching these settings makes the countries comparable within this source definition. The displayed level is not an annual percentage change, a forecast or evidence that one country caused a movement in another.

2. Apply the full reviewed period explicitly

Open Compare unemployment from the data catalogue. Check that Czechia, Germany and Poland remain selected and that the transformation is raw. To request the exact reviewed window, choose a custom period from 2016-10-01 through 2026-09-30, inclusive, and explicitly run it. Those bounds cover 120 source months for each country.

The generic MAX preset follows the application’s rolling date bound and can omit the first month of this fixed window. It is not a promise to include every reviewed observation. After changing the dates or selection, read the newly computed result and its actual period rather than assuming the earlier table still describes your controls.

3. Read gaps and revisions honestly

In the source capture accessed on 15 October 2026, this window contains 357 numeric observations and three missing September 2026 cells, one per country. The full grid therefore has 360 cells. September is not zero, and August must not be relabelled as September. Leave the gap visible instead of joining it with an invented observation.

This is a capture of the current revised history, not a collection of first publication vintages. Later source revisions may change earlier months or fill a missing cell. A difference between two reads needs its dates and provenance checked before it becomes a claim about an economic change.

4. Follow one observation into its evidence

Switch between graph and table without changing the applied query. Open Value provenance for a raw point to check its country, month, original decimal value, unit, quality and source. The graph, table, Passport and export refer to the same result; the Passport does not fetch a replacement value for the point you selected.

Separate the observation’s month from the dataset update, Datisto’s capture and ingestion, and result computation. Individual observation publication time is unknown in this capture. A dataset update timestamp cannot establish when a particular monthly figure first became public. A current access check likewise does not rewrite the result’s original computation clock.

1. Choose the question before reading the number

The inflation view offers three different measurements for Czechia, Germany and Poland. The annual HICP rate is the original percentage change against the same month of the previous year, reported by Eurostat as RCH_A. Datisto reads that separate source series; it does not reconstruct the annual rate from rounded index levels. Select this measure when your question concerns the pace of annual consumer-price change.

The I25 index instead reports a level with base year 2025 = 100. Price-level change over your selected period uses Datisto’s percentage-change method on that index. It is a change from a selected basis, not another source-reported annual rate or an annualised return. Changing the measurement changes the question, even when the countries and displayed months stay identical.

2. Keep the basket and exact period visible

Total HICP describes the source’s harmonised national consumer-price measurement. It is not your personal spending basket. Matching the classification and measure supports comparison within that definition, but a shared index base does not establish which country has higher absolute prices. Neither a lower annual rate nor a lower index alone proves that an identical household purchase costs less there.

The first visit prepares available default countries and a bounded period from actual coverage and reads the result. Apply a custom selection of one to three countries and both endpoint months, up to 120 months inclusive. For a requested-period change, the summary needs the exact usable positive starting observation and the exact ending observation. A later first usable graph point does not establish the requested starting month or the requested-period total.

3. Distinguish gaps, signed rates and revisions

An original annual rate can be negative or exactly zero. Preserve its sign and original decimal string; neither state means the observation is missing. An absent ending month remains unavailable. Do not insert zero, substitute the preceding month or join a gap with an invented value. The macro overview can identify an older numeric observation separately, with its own month; it does not relabel that observation as the requested month.

These series use the current revised history. The classification changed to ECOICOP2 from 2026, and earlier observations may be revised. Separate the measured month, dataset update, actual capture and ingestion, and result computation. Individual publication time may be unknown. A dataset update is not proof of the first publication of every monthly value, and a new access check does not turn a retained result into a new calculation.

4. Check each source and return to the selection

Czechia and Germany share a reviewed dataset within each measurement family. Poland uses a separate dataset for its index and another for its original annual rate. A mixed-country selection therefore offers separate CZ/DE and Poland source links. Check the selected dataset, methodology and current rights rather than assuming that one country’s source panel covers the other countries or every HICP measurement.

Both source links retain the same public selection: countries, measure and months. Return prepares those fields only; explicitly Apply to obtain a newly checked result. A shared selection link contains neither a saved result nor access permission. Source: Eurostat, PRC_HICP_MINR. The source panel contains the dataset DOI and review details. Selection, visualisation and this original explanation are by Datisto; Eurostat is not responsible for them.

5. Preserve the result you actually read

Switch between graph and table for the applied result, then open a point’s Value Passport. Check its country, original value, unit, period, quality, evidence and source method. A percentage-change point retains the underlying index observation as provenance; that source value has the index unit. Export reviews the same selected result and its manifest, rather than silently running another period or replacing the selected source.

1. Start with the exact GDP measure

This selection measures GDP at market prices using chain-linked volumes. Each percentage compares one quarter with the same quarter of the previous year, with seasonal and calendar adjustment. Read the source measure and unit before comparing countries. Datisto retains Eurostat’s reported annual change rather than calculating another growth rate from rounded GDP levels or treating the percentage as a price.

Real growth is a different measurement from nominal GDP in money, GDP per person or household income. This series also differs from change against the immediately preceding quarter, an annualised quarterly rate and a forecast. A larger growth percentage alone does not establish a larger economy, higher living standards or the cause of a country’s performance. Keep those questions separate from this result.

2. Read the whole quarter and preserve gaps

A quarterly observation covers its original calendar interval: January to March, April to June, July to September or October to December. Q2 2026 therefore belongs to April through June, even when you retrieve it in October. Inspect both boundaries in the Value Passport. The monthly macro overview has its own monthly observations; it cannot turn this quarterly measurement into three monthly GDP readings.

Negative growth and exact zero are valid numeric observations. Preserve the original sign and decimal spelling when checking an exported value. Missing data means something else: leave an absent quarter unavailable instead of inserting zero, carrying a preceding value forward or drawing an invented intermediate point. Match countries by the same original quarter, adjustment and measure before interpreting a difference.

3. Check the selected countries and current revisions

The reviewed selection contains Czechia, Germany and Poland over 38 quarters, from Q1 2017 through Q2 2026. That fixed window describes this available source capability; it does not promise future quarters, another country or every national-accounts measure. Select the countries and period you need, then check the applied query. A later expansion requires its own source and rights review.

The actual source capture on 15 October 2026 contained 114 numeric observations and 14 provisional flags on later German observations. Those counts describe that capture, not a permanent completeness guarantee. Provisional status stays attached to its original value. The history is the currently available revised history, so a later run may return revisions. It does not reconstruct what readers knew at the first publication of each quarter.

4. Separate the measured period from source clocks

The measured quarter, source dataset update, actual capture, ingestion and result computation answer different questions. For the retained 15 October source capture, the dataset update was 09:00 UTC and capture followed at 11:23:12.610 UTC. Neither clock proves the first publication time of an individual observation. An unavailable individual publication or revision clock remains explicitly unknown rather than borrowing the dataset’s update time.

Open the source panel to inspect NAMQ_10_GDP, its methodology and current permissions. Eurostat disseminates the statistics compiled under ESA 2010; the selection uses its original reported measure. Source: Eurostat, NAMQ_10_GDP. Datisto prepares this selection, visualisation and original English and Czech explanation. Eurostat is not responsible for those preparations, and the display does not imply its endorsement.

5. Verify and preserve the same result

Use graph and table to inspect the same applied result. Open a point’s Value Passport and check country, original decimal value, percentage unit, quarter, quality, evidence and method. Export reviews that result and its manifest. Compare its query and original rows with the displayed selection rather than assuming that a download silently refreshes the source or uses another period.

1. Identify the region and policy instrument

The deposit facility rate describes the Eurosystem’s overnight deposit facility policy instrument. Its unit is percent per annum: an annual rate convention attached to this overnight instrument. It is neither an annual percentage change nor a promise that a customer deposit earns that amount. Check the named instrument and unit before comparing it with any other percentage in the macro overview.

The source region is the euro area with changing composition, identified as U2. It is not Germany’s national rate, a Czech or Polish policy rate, or a fixed membership reconstruction. Keep it separate from the main refinancing operation, marginal lending facility, Euribor, bond yields and bank deposit or loan offers. Those measurements require their own definitions and source permissions.

2. Match the effective day rather than the announcement

Each original observation belongs to its dated UTC calendar day. A policy announcement can precede the day on which a new level applies. The ECB decision announced on 5 June 2025 took effect on 11 June. The retained source reads 2.25 on 5 June and 2 on 11 June. Moving the new value to the announcement day would change the historical record.

The reviewed window ends on 30 September 2026, when the original value is 2.5 percent per annum. Read that visible date together with the number. It establishes the level for the selected source day within this history, not today’s policy beyond the window. Neither the last value nor a stepped chart forecasts the next decision or promises a future rate path.

3. Preserve original daily values, zero and gaps

The captured API response contains 2465 original calendar-day observations, including unchanged weekends, despite a source title referring to dates of changes. Datisto does not create those weekend values by carrying a preceding rate forward. Graph steps present the returned daily levels. A source date at the beginning of the selected window is a boundary observation, not automatically a policy decision that day.

Negative rates and exact zero are valid source values. Preserve original decimal strings such as -0.5, 0 and 2.5 when checking provenance or downloaded rows. The source’s decimal metadata does not justify padding them with zeros. An absent value must remain a gap, with its reason visible; it cannot become a zero, interpolated estimate or silently substituted adjacent day.

4. Keep source clocks and reuse conditions distinct

Separate the effective day, actual capture, response Last-Modified, page update and result computation. The retained response and page metadata contain different update clocks, including older embedded chart metadata. Those clocks describe their respective source context. They do not establish when every daily observation was first published or revised. Individual publication and revision times therefore remain unknown, and current history is not a first-publication reconstruction.

Inspect the exact series and current rights in Sources. The original ECB statistics are available free of charge on the ECB website. Datisto prepares this selection, visualisation and original English and Czech explanation; no ECB endorsement is implied. This statistical source permission does not establish permission to reproduce ECB articles, images or logos, or to claim editorial approval.

1. Identify both target bounds and their unit

The FOMC target range has a lower bound and an upper bound. Read both for the same original date. The unit is percent of this policy target, not a percentage annual change, index or annualised investment return. Do not import the ECB source’s per-annum metadata into this separate Board measurement merely because both displays contain percentage signs.

The target interval differs from the effective federal funds rate, which measures market transactions. It is also separate from customer bank offers, a forecast and a probability path. Neither bound is an invented midpoint. Keep the two source series identifiable rather than replacing their interval with a single calculated number that could be mistaken for an observed market rate.

2. Read the fixed irregular history without filling days

This direct Board selection contains 55 original policy-date pairs, or 110 bound values, from 30 January 2020 through 17 September 2026. The final pair is 3.75–4 percent on that September date. It is neither a 30 September observation nor an assertion of today’s target beyond this window. The original document retains earlier context outside the selected analytical coverage.

The points include unchanged meetings, so each point is not necessarily a rate change. They do not form a daily series. Period end is unknown: the source provides a dated point, not a complete validity interval until another meeting. Connecting chart lines are presentation only; they do not supply intermediate observations, weekend values or permission to carry a bound forward.

3. Separate announcements, original decimals and missing bounds

Announcement and effective date can differ. The 17 September 2025 policy statement announced a 4–4.25 percent target, effective 18 September. The original chart document records 4 and 4.25 on 18 September. Keep that source point on its own date; moving it to the announcement date would alter the historical selection and its meaning.

Zero is a legitimate original lower bound. Preserve the original decimal strings without adding decimal places or replacing a missing value with zero. A missing bound cannot be reconstructed from its partner or an assumed spread. When both bounds exist, their exact numeric order must remain lower at or below upper, while their original spellings stay available for inspection.

4. Inspect the direct source and distinct clocks

Source: Board of Governors of the Federal Reserve System, Economy at a Glance – Policy Rate. This selection uses the Board’s own chart data, with a separately reviewed public-domain basis. It does not use FRED or New York Fed data. Datisto prepares the selection, visualisation and this original explanation; no Board endorsement, affiliation or editorial publication approval is implied.

Separate the original effective date, response update, actual capture and result computation. A response update may precede an effective policy date. None of these clocks establishes every observation’s first publication or individual revision time, which remain unknown here. The current revised history is not a first-publication reconstruction, and reopening a retained result does not make it a new source capture.

1. Read four maturities of the same nominal par curve

The Treasury view compares the original 2-year, 5-year, 10-year and 30-year Constant Maturity Treasury rates, or CMTs. Each maturity is a point read from the Treasury nominal par yield curve. A selected day therefore compares different maturities at one source date; the history chart compares each identified maturity across the dates actually supplied by Treasury.

These are fitted reference yields derived from indicative bid-side quotations. They are not executed bond prices, a particular bond’s coupon or the yield of a security that necessarily has exactly that remaining maturity. The four-point curve preserves the selected source points. Its connecting lines do not add observations for intermediate maturities or reproduce Treasury’s full fitting calculation.

2. Keep the annual yield convention explicit

The unit is percent on a nominal par, bond-equivalent basis. Treasury describes these as simply annualized yields consistent with semiannual coupon payments, using actual days on a 365- or 366-day basis. They are not compounded effective annual yields or APY. A displayed 5.00 percent remains that original source quote, rather than a promised five-percent investment profit.

Keep this nominal curve separate from real TIPS yields, an inflation breakeven calculation, a spot or zero-coupon curve and the federal funds policy target. Their economic meanings differ even when all are written as percentages. Treasury’s current nominal CMT method has a zero floor; zero is a valid observation. A missing cell is never a zero or a reason to invent a substitute.

3. Separate the source date from publication and capture

Read the date beside each yield before describing it as current. The initial capture contains source days from 2 January through 14 October 2026. A new download does not move the final observation to today. The midnight used to position a day on the chart is a date label; it does not claim that a market quote was measured or published at midnight.

Source update, actual capture and result creation are separate clocks. The feed update does not establish each value’s first publication or revision time; these remain unknown here. Treasury generally publishes after gathering that day’s quotations, with possible delays. Retained history is the source history available at capture, which may include revisions, rather than a reconstruction of what was first publicly known.

4. Inspect the selected coverage and original source

The complete original document contains fourteen maturities. This view deliberately selects four, with 197 source dates and 788 values in its initial capture. The other ten maturities remain in the source evidence; they are not silently included in the displayed result. No weekend, holiday or absent-day value is carried forward. An unavailable date stays unavailable instead of borrowing another day’s curve.

Use the Value Passport and Sources to inspect the original value, date, maturity, unit and method. Primary explanations are Treasury Yield Curve Methodology and Interest Rates – Frequently Asked Questions on the official Treasury website. This local development source is labelled licence unverified. Access to the public data is not a claim of verified redistribution rights, Treasury endorsement or editorial publication approval.

5. Check the common result and export scope

History, table and the selected-day curve use the same accepted four-series result. Choosing a day focuses its existing observations; it does not fetch a new dataset or change the original history selection. Inspect the exact point in its Passport when a rounded chart label is insufficient. The original decimal spelling remains available, including a genuine zero and trailing decimal places.

Export belongs to that same original result. Before downloading, check the reviewed dates, series and row count: focusing one curve day does not automatically restrict an export to four rows. Later reading and downloading still require current permissions. Neither retaining a result nor repeating the selection turns these historical reference yields into a live price, forecast or investment recommendation.

1. Confirm what one unit of your position represents

Available choices can include explicit Yahoo equity and ETF listings alongside the existing crypto references. One ETF unit means one share of that exact listing, not one unit of its underlying assets, cash or fund NAV. Yahoo Close includes the provider’s split adjustments; it is not Adj Close or a total-return price, and Datisto makes no additional adjustment. Legacy Yahoo assignments remain the same recorded source and are never silently moved to the new method. One report can select at most eight distinct price sources even when the eligible catalogue offers more.

Start with an existing open position and deliberately select a particular available price source. Confirm that one unit of your recorded quantity represents one unit priced by that source. A familiar ticker or similar asset name does not establish this relationship. For example, a broker balance described as a Bitcoin product is not automatically one Bitcoin, an ordinary share or a particular listed security.

The assignment is your explicit unit confirmation, not a verified broker or instrument match supplied by Datisto. Leave an uncertain position unmapped. Several separately recorded assets may use the same price source when their unit relationships are genuinely the same. This valuation is a separate report held in memory; it creates no trade, cash movement or new ledger entry and does not rewrite historical portfolio calculations.

To reuse assignments later, review and confirm Save assignments to your private portfolio. The saved document contains the unit relationships, not prices or a valuation report. A separate Load saved assignments action prepares your current draft; it does not run a market query. Check the loaded units and sources, then deliberately calculate a new report with current access. Unsaved edits and uncertain save requests remain temporary tab memory and can be lost after a reload.

Saving uses the reviewed document version and ledger identity. If either changed elsewhere, read the current state and review your retained choices again instead of overwriting another version. An unconfirmed response proves neither success nor cancellation: retry the same retained request when available. Clearing saved assignments is another explicit reviewed version; it does not delete transactions or change quantities. Closed positions keep their saved relationships until you deliberately remove them.

2. Keep quantity, price and both currencies visible

For a priced positive position, market value is its exact quantity multiplied by the selected original unit price. A real price of zero produces a real zero value. A missing price produces an unavailable value, never a zero estimate. Original price decimals remain available for inspection; a rounded label should not be treated as the full input to your calculation.

The accounting currency belongs to your recorded position and acquisition basis; the quote currency belongs to the price. They can differ. A USD quote values the quantity in USD even when the position was recorded in EUR. Native unrealized profit or loss is available only with a known basis in that same quote currency. No foreign-exchange conversion, percentage performance or investment return is inferred.

3. Read the source and its time before accepting the price

CoinGecko supplies a provider reference price assembled across its covered markets. Yahoo supplies a particular listing’s original quote.close field. Yahoo Close accounts for splits; Adj Close additionally accounts for dividends or distributions. Datisto applies no further price adjustment. Earlier v1 metadata used an unsupported unadjusted label, which remains recorded with an explicit correction rather than rewriting original inputs. These are different observations. Yahoo’s daily time identifies the source session bucket start, not the exact closing trade. Check the source date, capture time and price identity together; a recently downloaded old close does not become a new market observation.

The application uses maximum ages of one hour for CoinGecko and 96 hours for Yahoo in this selection. Those are application ceilings, not a verified market calendar or a promise of freshness. Weekends and holidays do not create substitute prices. Unknown source freshness remains unknown; passing an age ceiling alone does not establish a current executable market quote.

4. Read partial coverage before reading a total

Known position values are subtotalled by their quote currency. Cash remains a separate balance grouped by its own currency. The complete base-currency total appears only when every position is priced in the portfolio’s base currency and all cash uses that currency. Missing or mixed-currency inputs leave that total unavailable. A known subtotal is useful, but it must never be presented as the complete portfolio value.

Inspect each unavailable row: it can be unmapped, missing, too old or unsuitable for the selected source. The latest missing or rejected point is not replaced by an older numeric price. A relevant corporate action after the price can also prevent valuation: applying a pre-split price to a post-split quantity would mix incompatible units. Resolve that specific gap instead of silently adjusting either input.

5. Inspect and export the same calculation

Keep the calculation identifier and the source-price result identifier together. The report combines the selected position quantities, bases and cash with one accepted price result. Its Value Passport identifies the original source observation; the report’s multiplication is a separate derived value. Changing a mapping or portfolio input requires a corresponding fresh calculation rather than reusing an unrelated price explanation.

Before using a JSON or CSV export, check its calculation, source-price result, coverage and currency labels against the visible report. Reading and downloading require current source permissions; a previous selection does not grant permanent rights. Treat this memory-only report as a private current valuation aid, not a stored accounting event, broker reconciliation, historical performance series or recommendation to trade.

6. Compare the whole valuation with your own targets

In the single-currency valuation, open My target allocation. Enter your own weight for every instrument and cash group, including explicit zero where intended. Weights must sum to exactly 100%; no target or remainder is recommended or filled automatically. Review applies the draft to that same original valuation. Edit targets returns to your retained choices. A different ledger, mapping, currency, fixing or source result requires a corresponding fresh review.

Drift is current weight minus target in percentage points. Gross adjustment is target amount minus current amount in the presentation currency. The calculation uses unrounded exposures, and exact adjustments cancel before independent output rounding. It requires complete coverage, nonnegative exposures and a positive whole value, with at most 50 groups. The known portion of an incomplete portfolio cannot stand in for the whole. Fees, taxes, trade quantities and execution are not included. This model creates no orders or ledger entries.

Target model JSON retains the target choices, complete result and original valuation/source artifact. CSV is a table of the same complete model. Both downloads recheck current access and the original ledger/input identity. Unsaved target drafts and calculated models remain private tab memory; a download is not a server Save or a restorable ledger backup. To keep the own weights, separately review and confirm Save targets. Load saved targets explicitly restores those weights into the draft without running a price query or a model calculation. Recheck an expired original input before using the model again.

7. Choose a broader crypto source explicitly

When an admitted broad CoinGecko capture is available, search its assets by name and exact source coin ID. Select that separately labelled V4 USD reference and confirm your recorded unit relationship. A ticker alone never assigns a price, and existing BTC/ETH or equity assignments do not silently switch source. The available list is the actual captured cohort of up to 250 assets, not every asset ever registered.

A report can combine supported equity sources and crypto assignments while keeping the limit of eight selected price series. The complete crypto capture is retained as evidence; its companion rows are not additional portfolio positions. A genuine zero price remains zero. An absent asset, a null price, a source beyond the one-hour crypto application age limit or a point later than your valuation cutoff stays unavailable. A missing companion percentage alone need not invalidate an otherwise admitted original price. No older cohort supplies a departed member.

Review and Save keep the exact source assignment in your private account, without storing a price or changing the ledger. An unavailable saved assignment remains visible and removable. Whole input JSON and CSV retain the full original source, exact selected prices and any CNB inputs; the private valuation export preserves this same evidence and calculation. Source refresh limits and current permissions still apply.

Full crypto identities and page prices

The full crypto identity catalogue loads automatically on opening. Search checks the entire captured identity list by source ID, name and symbol; repeated symbols remain separate assets. Results use a deterministic source-ID order and pages of 100. The catalogue contains identities, not a price for every asset or a whole-market ranking. An exact requested ID that is absent is never replaced with the first record.

After the catalogue loads, prices for the displayed page load automatically. Changing pages or search reads only those IDs, with search waiting briefly while you type. Source null, a missing price field and an omitted source row are different states, and a reported zero remains zero. Inspect the selected exact identity and its original fields, source update time, page capture and catalogue capture separately. Identity-list coverage does not establish quote coverage, a common all-market clock, verified versions or an executable trade price. An exhausted source budget leaves identities available and prices honestly unavailable.

Download page JSON or CSV retains every requested identity in order, including unavailable prices and records not returned. The full identity file is separate. Files preserve normalized source fields and original capture hashes; they do not claim to reproduce upstream raw bytes. Download first checks current access to the same retained cache identity; expiry or a changed source refuses the original file instead of substituting new prices. The separate up-to-250 overview and its existing original analyses keep their own source scope.

Open an exact asset and continue to History and comparison. Selecting an identity or changing the 7, 30 or 90-day period automatically reads its bounded original history; Refresh repeats the read. Prices, market caps and rolling 24h volumes retain all their original timestamped points, units, nulls and missing fields. Switching the displayed series or table page never crops the whole JSON or CSV. Provider sampling is not a daily Close series; source point time, history capture and identity-list capture remain separate.

Pin two to six exact IDs in the catalogue table or asset detail. The ordered selection stays in this page’s memory across search and pagination; move or remove members explicitly. A valid two-to-six identity selection and period load automatically; use the control to refresh the same request. Every member remains in the result. If one history is unavailable, the comparison is incomplete and no smaller substitute chart is used. A complete comparison uses only exactly shared original timestamps and a first common point with all prices positive; base 100 expresses relative USD price movement, not total return. Excluded timestamps remain visible and break chart lines. Whole files preserve every selected source, unavailable member and exclusion, independent of the open table page. The selection stays in this browser tab.

2. Choose a particular source, pair and currency

A market price belongs to an identified instrument and a specific observation. Open Market prices and choose the source group you intend to inspect. A provider reference, a venue trade and a completed daily close describe different things. Changing the source selector prepares a choice; use Open selected source to apply it. Read the active source description before interpreting the cards or the chart below it.

Coinbase Exchange BTC/USD and ETH/USD daily closes use USD as their quoted currency. Binance BTC/USDT and ETH/USDT use the USDT stablecoin. Similar numerical labels do not establish currency equality, a conversion rate or an arbitrage opportunity. CoinGecko references combine a provider’s covered markets and are not an executable quote on either exchange. Keep the quote unit and source family visible whenever you compare observations.

3. Distinguish a daily bucket from a trade time

The Coinbase daily series selects the original closing price from a completed UTC bucket. Its date coordinate is midnight at the beginning of that source day. The last trade inside the bucket supplies the close, but its exact timestamp is not supplied by this series. An unfinished current bucket can appear in the downloaded response; it is retained as raw context and excluded from completed daily prices.

Yahoo daily coordinates instead identify the start of a particular listing’s source session bucket. A Binance last-trade series has its own trade timestamp. Those clocks must remain separate from capture, ingest and access-verification times. A recent download of yesterday’s completed close does not turn it into today’s last trade. Unknown publication time or freshness remains unknown; the interface does not infer either from the server clock.

4. Inspect original decimals and honest gaps

The original numeric price travels as an exact validated decimal string. Use the table, Value Passport or export when the full input matters; a compact chart-axis label is only a display aid. A real zero is distinct from a missing value. Replacing a missing or rejected latest observation with an older price would answer another question, so inspect the stated unavailable condition rather than treating every card as a current executable quote.

History coverage reports what was actually stored for that series. A source can omit a no-trade interval or return buckets outside the requested window. Neither case authorizes inventing a close. The selected Coinbase capture contains ninety completed days per pair; that measured sample does not promise unlimited history, all markets or future completeness. Check the requested window and each series’ real coverage before drawing conclusions about a missing day.

5. Use the same result in every presentation

The graph and table are two presentations of one admitted result. Switching presentation should not replace the source or quietly calculate another recipe. Open a point’s Value Passport to inspect its original observation, source method, unit and time meanings. Keep its result identity when downloading the underlying JSON or CSV. The file should describe the same observations and coverage that you reviewed on screen, including any explicit gaps.

Access is checked when a result is read or downloaded. A saved selection, an old browser tab or a previously visible chart does not grant permanent permission to source data. If a source becomes unavailable, select another source deliberately and inspect its different identity. Do not assume that a second provider reproduces the first provider’s method, values, time coverage or redistribution rights merely because the asset name matches.

6. Keep price inspection separate from investment return

An original price is an input to analysis, not a recommendation to trade. A change between two prices does not include your deposits, quantities, acquisition fees, dividends or currency conversions. The separate portfolio workflow requires explicit own-unit assignments and shows its coverage. It must not guess that a broker product represents one Bitcoin or one share. A source choice in Market prices does not create a transaction or automatically assign a portfolio position.

Before reusing a chart or report, write down which source family, pair, currency, price type and period answer your question. Preserve the original result when another person needs to inspect your evidence. Local development availability is separate from public licensing and release approval. This guide explains the implemented data workflow; it supplies neither an execution service nor a claim that a historic observation can be traded at its displayed price.

7. Read the wider crypto source explicitly

Choose Broad selection for a V4 capture of up to250 assets, then read it explicitly. The displayed source label and actual returned count belong to the applied result. The old twenty-asset classified source remains a separate choice. If the wider original is unavailable, its status offers an explicit read of the old source; neither prices nor a claimed250-member result are invented. Changing the source selector alone does not replace an already displayed original.

The free source has shared request limits and backoff. In keyless operation, API and worker can reuse the same original response for twenty minutes, preserving its capture and hash. A genuinely new acquisition can receive twenty-minute application freshness; an already stored fifteen-minute receipt keeps its original expiry, and first projecting an older reused original conservatively keeps fifteen minutes. Expired captures are labelled last known. These limits do not supply live pricing, a new source update or historical market membership.

1. Choose one instrument and its original market

Open the market candle screen from Market prices. Choose Bitcoin or Ethereum on the explicitly named Coinbase Exchange USD market. Prepare thirty or ninety completed UTC days, or enter your own from and through dates for one to ninety days. The through date is included in the form; the query ends at the following UTC midnight, exclusively. Open the complete selection deliberately. Editing an instrument or date keeps the existing chart and result until you open a valid new selection. Read the applied source and instrument beside the displayed data before interpreting a chart. This first workflow covers original completed daily intervals in a bounded source window, rather than every exchange, asset or trading frequency.

USD is the source quote currency for these two products. A Binance USDT candle uses a different quote asset and venue; a CoinGecko reference point has a different method. Neither becomes the same Coinbase candle merely because the numbers look similar. The application keeps these identities separate. Selecting a display currency elsewhere does not convert an original price or this market's volume.

2. Read the whole completed UTC interval

Open is the first traded price within the source interval and close is its last traded price. High and low are the interval's reported price extremes. They describe historical activity on this particular market, not a current bid, ask or executable offer. A candle body connects open and close; its wick shows the low-to-high range. The chart is a visual presentation of these original values.

A daily interval starts at 00:00 UTC, including that boundary, and ends immediately before the next 00:00 UTC boundary. Its date labels the start, not the exact timestamp of the final trade. Only completed intervals are admitted. The current unfinished day stays out of completed history. Datisto capture, ingestion and any known publication are separate clocks; an unknown publication time or provider freshness remains explicitly unknown.

3. Keep price and volume units distinct

All four price fields use USD per unit of the selected asset. Volume measures the amount of the base asset traded during that same interval: BTC for BTC-USD and ETH for ETH-USD. It is not USD turnover, market capitalization, a rolling twenty-four-hour counter or total activity across all venues. A larger numeric ETH volume cannot be compared directly with BTC volume as though the units were identical.

Source decimal strings are retained, including meaningful precision in the original document. A chart or localized number may use a shorter display format; inspect the value or original export when exact digits matter. An actual numeric zero is a value. An absent interval, unavailable value and invalid source response are different states. The application does not replace any of them with an invented zero or estimated trade.

4. Switch presentation without changing the calculation

Candles, the closing-price line, base-volume bars and the exact observation table are presentations of one admitted result. Switching presentation should not fetch a separate quote or silently choose another source. Check the result identity and covered interval. The table preserves the original fields and interval boundaries; the Value Passport connects a selected observation to its original source, capture, method and current access verification.

The requested window is a limit, not a completeness guarantee. Invalid dates, reversed or oversized windows and the unfinished current or future day cannot be opened. A valid older window can contain no stored observations; the empty result states that absence and keeps its requested dates. Source days without data remain gaps; the previous close is not copied forward to create a candle. Thirty candle dates also do not establish a thirty-day return: that needs a separately defined starting price and calculation. A retracted or suspect latest observation cannot borrow an older candle to appear current. Weekly and monthly aggregation require a separate method and are not supplied by relabelling this original daily feed.

5. Export originals and interpret corrections carefully

An original JSON or CSV download refers to the same result and includes the OHLC fields, base volume, units and interval context. Download permission is checked when reading it; an expired review may need fresh verification. A cached or previously saved result is not permanent permission to reuse its source. Public redistribution and provider licensing are separate from this local development workflow.

A provider may correct a high or volume while leaving the close unchanged. That is still a revision of the whole candle, and an earlier accepted result keeps its original inputs rather than being silently rewritten. These candles describe prices and activity, not an investment recommendation, portfolio return, dividend-adjusted performance or proof of why a price moved. Use the stated source and interval before drawing a conclusion.

1. Choose the contract and the historical window

Open Funding from the market overviews. Choose BTCUSDT, ETHUSDT or both and explicitly apply that choice. Each is a Binance USDⓈ-M perpetual contract with USDT as quote and margin asset. Bitcoin held in a wallet and the BTCUSDT perpetual contract have different identities; this screen does not report a spot purchase price or the value of your portfolio.

The result contains settlements actually retained from the source. Use the period controls to create a new result for a different window. A missing point stays missing. The oldest stored point is the beginning of available coverage, not proof of the first settlement in the contract’s lifetime.

2. Read the sign and distinguish percent from the source ratio

The table presents the source rate as a percentage: an original ratio of 0.0001 is exactly 0.01%. This is a decimal shift, not a newly estimated fee. Passport, the graph and original exports retain the source ratio per settlement. Check the unit before copying a value into another calculation.

Under the venue’s convention, a positive funding rate means long positions pay short positions; a negative rate reverses that direction. It does not describe whether the market price rose or fell. Your own cash payment also depends on the eligible position and venue rules, so the displayed historical rate is not a cash entry in your ledger.

3. Keep settlement time separate from capture and interval length

Each row keeps the original settlement timestamp, including milliseconds. Datisto’s capture time says when the source document was received. Neither timestamp establishes when the exchange first published that row. Open a rate’s Passport to inspect its original value, source and recorded clocks.

Funding intervals can change. A currently captured interval setting cannot establish the duration used by every historical rate. A gap between two reported timestamps is not, by itself, a verified charged period. This workflow leaves historical interval duration unknown and does not multiply a rate into an annualized yield or sum rates into a payment.

4. Verify, preserve and return to the same analysis

The graph, paged table and exports use one admitted result. Paging changes only the visible rows. JSON and CSV download the original result with its declared units and provenance; they do not silently turn source ratios into displayed percentages. An expired review may require fresh verification before another read.

1. Select the contract and its source

The COIN-M overview concerns Binance BTCUSD_PERP and ETHUSD_PERP perpetual contracts. These contracts are quoted in USD and collateralized in BTC or ETH. They are separate from spot assets and from Binance USDⓈ-M USDT contracts in the funding overview. The venue and contract family are part of the meaning of each value.

Choose BTC, ETH or both, then apply the prepared selection explicitly. Before applying it, the admitted result remains the previous one. The available history is the range actually captured from the source; it is not the full history of the instrument or the total across all exchanges.

2. Keep contract counts separate from coin quantities

The main value is the original sumOpenInterest contract count. The table also preserves sumOpenInterestValue in the contract’s base asset: BTC for BTCUSD and ETH for ETHUSD. These are two source-reported quantities at the same source point, with different units. A count of contracts is not a number of Bitcoin, and BTC and ETH quantities cannot be added together as one coin balance.

The chart displays contract counts. Its lines retain their own contract labels and do not imply equal dollar exposure. Current contract metadata alone does not prove the multiplier effective at every historical point. This workflow therefore leaves historical USD notional unavailable instead of manufacturing it from an unverified historical conversion.

3. Interpret the reported time and limits

The source timestamp identifies the reported period end for a daily open-interest observation. It does not mean the day’s traded volume, an average over the whole day or a publication clock. Datisto’s capture clock separately records when the complete source evidence became available. Open Passport to inspect the exact value, original coin quantity and recorded source identity.

A larger count says that more contracts remain open under the source’s definition. It does not by itself identify new long demand, new short demand, leverage, a liquidation level or a future price direction. Unknown or unavailable values remain distinct from a genuine reported zero. Later refreshes may add observations; the stored capture is not a guarantee of realtime availability.

1. Identify the index and its unit

The US inflation view uses the original BLS series CUUR0000SA0: Consumer Price Index for All Urban Consumers, US city average, all items, not seasonally adjusted. Its reference base is 1982–84=100. This is a monthly index level, not a percentage, core CPI, chained CPI or a measure of every household’s own spending.

The prepared actual overview loads on opening. Read the selected monthly range and source before interpreting a value. The base of 100 represents the average reference-period price level; it is not an annual inflation target. The separate European HICP view uses another statistical definition and source, so the two indices are not interchangeable.

2. Compare exact calendar-month endpoints

Monthly change, MoM, is (current index / previous calendar month’s index − 1) × 100. Yearly change, YoY, uses the same calendar month one year earlier instead: (current index / index twelve months earlier − 1) × 100. YoY is a twelve-month endpoint comparison, not the average of twelve monthly rates or a comparison of annual averages.

Both changes in this view are calculated from the published, not seasonally adjusted index. In particular, NSA MoM must not be presented as the seasonally adjusted monthly headline commonly quoted in news. Percentages are derived from the exact published index strings; a rounded display does not replace those inputs, and these calculations do not claim access to unpublished BLS precision.

3. Keep a genuine source gap visible

The captured source explicitly reports October2025 as unavailable, with footnote X explaining the 2025 lapse in appropriations. The original dash becomes a missing index with its explanation. It is not zero and is not replaced by September, November, an interpolation or an annual average.

A missing endpoint also makes the corresponding derived rate unavailable. October2025 has no index or derived change; November2025 has no MoM because its previous calendar month is missing. A November-to-September comparison would span two months and cannot substitute for November MoM. Future months and unreported endpoints likewise remain unavailable, even if another displayed rate is valid.

4. Separate the measurement month from release and capture

September identifies the month measured by the index. Its UTC month boundaries organise the analysis; they are not a release timestamp. Datisto’s capture clock records when the original source response was received. The API’s optional latest marker indicates the newest source row, and its responseTime describes request duration. Neither establishes when BLS first published that monthly value.

This source response supplies no per-value release or revision clock, so release time remains explicitly unknown. BLS describes unadjusted CPI-U as final when issued but can correct errors; that general policy does not establish an original-release vintage for this capture or promise that no correction can occur. A stored analysis preserves its captured inputs rather than silently taking a later source revision.

5. Inspect the inputs and preserve one result

Open an index’s Passport for its original value, source month and footnote. For a MoM or YoY cell, open the paired-endpoint Passport: inspect the current index and the exact previous-month or previous-year index together, including their original source evidence. An unavailable endpoint remains visible as unavailable; the derived percentage does not acquire an invented monthly observation.

1. Open the eight-maturity view

The dedicated Treasury curve view selects 1M, 3M, 6M, 1Y, 2Y, 5Y, 10Y and 30Y from the original US Treasury nominal par constant-maturity source. Month and year labels identify time to maturity; 1M is not one year. These are eight selected source maturities, not the complete fourteen-tenor feed. The separate earlier four-maturity view and its saved originals remain available.

The prepared original result loads automatically; check its source dates and coverage. This version concerns the captured 2026 history. Its first local capture contains 197 source days, 2 January–14 October 2026, with eight original values per day. Those dates describe the available capture rather than a promise of complete instrument history or a continuously updated quote.

2. Compare maturities within the exact same day

Use the exact source-day selector to inspect the cross-section. The horizontal scale represents maturity in years, with the month labels retaining their own shorter durations. The plotted values all belong to the chosen source day. A connecting line helps show their shape; it does not supply an original yield for a maturity between two reported points.

Treasury produces its own par CMT estimates using indicative market inputs and its curve methodology. Datisto preserves the selected published values and does not fit another curve, create an unreported tenor or substitute a bond trade price. A dated upward, flat or inverted shape compares these particular nominal par yields; it is not a forecast or a calculation of your investment return.

3. Read the yield unit and leave gaps intact

Values are nominal par CMT yields in bond-equivalent percent. Read 4.11 as 4.11%, not as a price, a decimal return of 4.11 or compounded APY. The quotation convention differs from real/TIPS yields, spot or zero-coupon rates, a policy rate and a realised holding-period return. The source’s two-place decimal spelling remains available in Passport and the original downloads.

Only original source days appear in the history. Selecting an absent weekend or holiday does not silently load the previous business day. A source N/A or XML missing marker stays a gap, while an explicitly reported zero stays zero. Do not read a missing point as no yield or estimate it from adjacent maturities or neighbouring days.

4. Separate source date, update and capture clocks

The printed source date tells you which day the yield belongs to. Midnight UTC is its chart coordinate, not proof that the market was measured at midnight. The feed’s update metadata and Datisto’s capture time are separate clocks. Neither establishes the first publication time of each individual yield or a verified intraday quote time.

Per-value publication, provider revision and finality remain unknown in this source contract. A later source capture can contain corrected history. Inspect the recorded capture and freshness before using the result; a source-date selection does not refresh the feed or make older captured values current.

5. Verify a point and preserve the whole original

Open Value Passport from a maturity value or an original historical point to inspect the exact source value, tenor, unit and evidence. Curve, history chart and table use the same verified result. Changing the displayed day, paging the table or hiding a legend series affects presentation, not the original query scope or the stored source values.

6. Calculate the difference between two maturities

The separate Treasury spread workspace calculates the original ten-year nominal par yield minus the two-year yield, with ten-year minus three-month as the other supported pair. Choose the pair and bounded 2026 date window, then Calculate explicitly. Editing these fields is only a draft. The difference is measured in percentage points; the two original yields remain percentages. Positive, exactly zero and negative differences describe those same-day quotations, without supplying a recession probability, investment return or forecast.

1. Understand what this real yield measures

This view uses the US Treasury’s original real par constant-maturity yields derived from Treasury Inflation-Protected Securities, or TIPS. Treasury estimates its real curve from indicative secondary-market quotations and reports values at fixed maturities. A constant-maturity point need not equal the yield of one identified bond with exactly that remaining lifetime.

Datisto does not calculate this real yield by subtracting today’s CPI rate from a nominal Treasury yield. The source value is not breakeven inflation, a forecast of prices or the realised purchasing-power return of your own bond investment. Nominal and real series retain different units and source identities.

2. Compare the five reported maturities on an exact day

The real source reports 5Y, 7Y, 10Y, 20Y and 30Y. The prepared result loads automatically; select the exact source day for the cross-section. There is no reported short real maturity in this selection to substitute for a nominal one-month or one-year point. A line connecting the five visible points helps read their shape; Datisto supplies no new yield between those maturities.

The initial real capture contains 198 source days, 2 January–15 October 2026, and 990 original values. The earlier retained nominal capture ends on 14 October. Their different coverage is explicit: the latest real day cannot be paired with the previous nominal day as though both came from the same date. This view performs no such pairing or breakeven calculation.

3. Read real bond-equivalent percent and keep gaps

The unit is real par bond-equivalent percent, quoted on the source’s simple annualized convention for semiannual coupons. It is not a compounded APY, a bond price or a guaranteed annual return. Original two-place decimal strings remain available through Passport and the downloads; display rounding does not change the captured source value.

Only reported source dates are present. An absent weekend or holiday does not silently reuse the previous business day. A source missing marker stays unavailable rather than becoming zero or an interpolated value. The parser can preserve a genuinely reported negative or zero; the initial 990 actual cells are all positive, so this capture is not evidence of actual negative or zero observations.

4. Separate the source day, update and capture clocks

The printed source day identifies which day a yield belongs to. Its midnight UTC chart coordinate does not establish an intraday market measurement. Feed and entry update timestamps describe source metadata, while Datisto’s separate capture clock records when the full response was received. Neither proves an individual point’s first publication time.

Per-value publication, provider revision, period end and finality remain unknown in this source contract. A later capture may contain corrected history. Check capture and freshness separately from the selected source day; selecting a newer-looking day does not refresh the data or turn a stored estimate into a current executable quote.

5. Inspect a point and preserve the complete original

Open Value Passport from a maturity or an original historical point to inspect its exact value, real-yield unit, source field and evidence. Curve, history chart and table share one verified result. Switching the day, paging the table or hiding a legend series changes presentation, not the original query scope or the provenance of another point.

1. Understand the market spread

Inflation compensation, often called breakeven inflation, is the nominal Treasury yield minus the TIPS real yield at a comparable maturity. The spread also reflects inflation risk and relative TIPS liquidity. It is an imperfect proxy for inflation expectations; this view does not estimate those premiums or turn the spread into a forecast of future CPI.

Datisto calculates its own named difference of the two original Treasury par quotations. It is not a retrieved FRED observation, a fitted zero-coupon forward rate, a subtraction of current CPI or the realised return of a bond investment. The original nominal and real source families remain separately identifiable.

2. Pair the exact day and maturity

The calculation uses 5Y, 10Y and 30Y, the three maturities present in both admitted families. Each cell pairs only its exact source day and maturity. There is no nearest-day substitution, forward fill or interpolation. The real 7Y and 20Y points have no matching series in the selected nominal eight and do not become additional breakeven points.

The initial retained sources contain 197 nominal days through 14 October 2026 and 198 real days through 15 October. Their union produces 594 day-and-maturity cells: 591 available pairs and three unavailable cells on 15 October because that nominal day is absent. The latest complete comparison is therefore 14 October; choosing 15 October keeps the genuine gap visible.

3. Read percentage points and basis points

Subtract the two original percent quotations directly. A nominal 5.00 and real 2.64 give 2.36 percentage points, equal to 236 basis points. One percentage point is 100 basis points. This is a difference, not a percentage change or another annualisation. Exact integer arithmetic preserves the source hundredths without calculation rounding.

A negative or zero difference remains a signed result or an actual zero. A missing, withheld or unusable original point produces an unavailable cell with a reason. It never becomes zero. The complete original query retains all eight nominal and all five real series, including the seven maturities unused in these three paired calculations.

4. Inspect both source clocks and originals

A matching source day does not mean both responses were captured together. The retained nominal response was captured on 15 October and the real response on 16 October, each with its own full original checksum and timestamp. Individual publication and revision clocks remain unknown. A midnight UTC coordinate identifies the source day without asserting an intraday observation or first publication.

Open Value Passport from a paired cell to inspect the calculation and both original point proofs in one verified read. On a gap day, the available real original can still be inspected and the absent nominal input stays explicit. The seven unused source series retain their ordinary raw point passports. Current access to both datasets is checked again.

5. Preserve the complete analysis

The initial full result has 2566 original source rows and 594 analytical cells. CSV preserves the analytical rows and both endpoint identities, original values, units, capture clocks and evidence hashes. JSON also retains every original row from all thirteen series and the full derived document. Switching the visible day, unit or history range does not narrow the original saved or downloaded result.

1. Understand this derived ETH/BTC value

Datisto divides the original Coinbase Exchange ETH-USD daily close by the original BTC-USD daily close. The USD units cancel, leaving BTC per one ETH. This is a named calculation from two USD markets, not an executed trade or an executable quote in a directly traded ETH-BTC market. Neither spot spread, fees nor trading slippage are included.

2. Match the exact completed UTC bucket

Both inputs must describe the same 24-hour UTC bucket, the same venue, original unadjusted prices and the correct Bitcoin and Ethereum identities. Each close is the source-reported close within its own daily bucket; matching buckets do not prove that the last trades occurred simultaneously. Missing one exact input leaves a visible gap, without borrowing the previous day or another exchange.

The initial interval uses the latest common stored ETH and BTC coverage, bounded to at most 90 days. Each ratio cell needs both original daily closes; missing values are not filled. Latest value, history and table share one result. Presentation changes do not switch the source.

3. Read precision and unavailable values

The original decimal strings enter exact rational division. Datisto rounds once to at most 18 decimal places using HALF_UP and removes trailing zeros. The rounding indicator is true when a nonzero remainder was discarded at that precision; a displayed value is not presented as infinitely exact. Original prices retain every supplied digit.

A genuinely usable zero ETH close gives ratio zero when the BTC denominator is positive. A zero BTC denominator, missing original, unusable quality or out-of-budget ratio is unavailable with a reason. An unavailable value never becomes zero. The calculation does not infer a future price or an investment recommendation.

4. Inspect the two originals and capture clock

Open Value Passport from a ratio cell to inspect both original OHLCV points in one verified read: Ether is the numerator and Bitcoin the denominator. The retained source bundle has its own capture clock and checksum. That bundle clock is not the time of both closing trades or a proven first publication. Individual source publication and revision clocks remain unknown.

1. Identify the asset and price source

This Ethereum detail uses the original ETH-USD market on Coinbase Exchange. Its unit is USD per one ETH. The price is an unadjusted daily close within the source-reported UTC candle; it is not a current executable quote, an aggregated CoinGecko reference, a USDT price or a forecast. Bitcoin, a token with a similar symbol and another venue cannot silently replace this source.

2. Read close history, candles and volume together

Volume is reported in ETH units for that daily bucket, not USD turnover or a rolling 24-hour figure. A missing candle stays a gap rather than borrowing the previous close or inventing zero volume. A withheld original remains unavailable even when an older usable price exists. Other saved source windows retain their exact original boundaries.

3. Inspect source time and the original point

The last displayed price has its source day, exact bucket bounds and source freshness. The capture clock belongs to the retained source bundle and does not establish publication or the instant of the closing trade. Individual publication and revision clocks remain unknown. Open Value Passport to inspect the original decimal strings, OHLCV fields, asset identity, evidence and current source access.

1. Understand the index-point unit

These records are Yahoo Finance instruments classified INDEX. They report the original daily quote.close field in index points, under a separate cash-index metric. The provider’s USD currency basis does not turn a level into a price in USD per unit or a quantity you can buy. SPY and QQQ remain separate ETF listing prices.

Each index has its own native base, composition and history. A higher point level does not show a better return or a larger market. Select one index for the historical graph; the other cards and full table remain part of the same result. This view does not normalise levels to a common base or calculate a total-return benchmark.

2. Read the source session and capture separately

The original timestamp identifies a daily source session-bucket start in America/New_York. It is retained at its original instant rather than snapped to midnight UTC. It establishes neither the time of a closing trade nor the publication time of the level. Source dates are displayed using the exchange timezone.

The displayed capture is the stored source-bundle timestamp. The three original HTTP captures retain their individual clocks in the original evidence. Equal latest session dates establish neither a simultaneous capture nor quote. The captured current regular session is excluded; unknown publication, freshness and finality remain explicit. A missing original close is a gap, not zero, and skipped sessions are not invented.

3. Use one result across cards, graph and table

The overview reads the explicitly selected known index series in one bounded raw query. The ordinary view includes all three indices. Each card shows that series’ latest original point with its own date and capture; it does not silently replace an unavailable latest value with an earlier usable level.

Selecting a graph index changes only presentation. The full table and download keep every originally selected index and source row in the same result. Open a point’s Passport to inspect its exact original decimal string, source series, evidence and quality. Rounded labels never change the source value.

4. Retain the original recipe and provenance

CSV retains the full ordinary source table in 14 columns; JSON retains the complete typed result and manifest. Switching the visible index or table page does not narrow either download. Current download rights are verified for the same result before the file is handed to the browser.

5. Compare from one shared source session

Compare indices prepares the three supported actual indices and their available common period, at most 100 calendar days. Later choose two or three indices and custom From and Through-inclusive UTC dates, then Calculate. Original index-point levels remain in their separate overview.

The comparison starts at the earliest exact source-session coordinate inside the applied interval where every selected index has a usable positive original level. Every selected index equals 100 at this one shared session. Later values follow original level divided by that index’s original basis level, multiplied by 100. Normalized values are rounded to 18 decimal places with HALF_EVEN; exact original levels remain unchanged. The requested calendar start is not silently presented as a traded session.

An original input before a later common basis stays retained without a normalized value. Absent or unusable inputs remain gaps; there is no outside-window or nearest-session replacement. Without a valid common basis the comparison is unavailable. The shared basis makes price-level movement comparable; dividends, total return, currency conversion and forecasts are outside this calculation.

1. Keep the source pair and units explicit

This page uses the separate original Coinbase Exchange SOL-USD source. Open, high, low and close are USD per SOL; volume is the original SOL quantity in that daily bucket. It is not USD turnover or a count of trades.

These venue-specific candles remain separate from CoinGecko aggregate references and from BTC-USD or ETH-USD. No other source replaces a missing SOL candle. The history does not establish staking income, network activity or a forecast.

2. Select completed UTC days

The initial live window is prepared from the actual stored SOL coverage, up to 90 calendar days. The last-30-day shortcut prepares a draft from the latest stored source day. It is not a claim that prices are current today or that every day is present.

The From and Through inputs include both selected dates. Apply window explicitly starts a new result; typing preserves the accepted result. The stored query uses a half-open interval from the first midnight UTC to the midnight after the final date. Each original candle retains its exact start and exclusive end.

3. Separate the source bucket from knowledge clocks

A candle timestamp is the daily bucket start, not the closing-trade or publication instant. The source-bundle capture is shown separately. Individual publication, provider revision and finality are unknown; an old source close does not become a live quotation because you opened it now.

Absent candles remain gaps and withheld values remain concealed. Missing is not zero. Open the actual row’s Passport to inspect its original OHLC fields, SOL volume and evidence; a source-date match does not establish a simultaneous cross-market quote.

Chart, volume and the same data in exports

Candles, close, volume and table display one bounded result. CSV preserves the original 26-column OHLCV export and JSON retains the full result. Switching views does not recalculate prices, change the result or restrict its export to a displayed table page.

1. Keep the derived ratio separate from a traded pair

The numerator is the original SOL-USD close and the denominator is the original BTC-USD close. Their quotient expresses BTC per 1 SOL. This is not the price of an executed SOL-BTC trade, a simultaneous quote, a percentage return or a forecast.

Both inputs must describe the identical completed daily UTC bucket. Matching bucket boundaries establish the calculation basis, not simultaneous provider requests or publication. Each original OHLCV context and source-bundle capture remains separate.

2. Read the whole source window, including gaps

The initial window is prepared from actual stored SOL coverage, at most 90 calendar days. It is not shortened to the intersection of both histories. The retained July-to-October source window includes a final SOL day without a matching BTC candle; that missing denominator remains visible.

The untouched day selector may start at the latest available pair. Choosing a later gap keeps that exact day and its real SOL input. No prior BTC close is carried forward, no other venue is substituted and no missing value becomes zero. The displayed latest source days can therefore differ.

3. Inspect the recorded ratio and both original inputs

The server divides validated decimal strings and records up to 18 decimal places using the named HALF_UP policy. The rounded flag distinguishes a rounded quotient from an exact finite quotient. The browser only positions those recorded values in the graph; original dollar closes are not rounded by the calculation.

Open the paired Passport from an actual source observation. It verifies the ratio cell, both available originals and their separate capture clocks in one current read frame. On a missing-counterpart day it retains the real endpoint and explicit absence, while current permissions for both exact datasets remain required. Publication, provider revision and finality stay unknown.

1. Keep the venue and original units

The page uses original unadjusted Coinbase Exchange BTC-USD daily candles. Open, high, low and close are USD per BTC; volume is BTC quantity in that exact daily bucket. It is neither dollar turnover nor a rolling24-hour total.

CoinGecko aggregate references and Binance USDT instruments remain separate sources. A missing Coinbase candle is not replaced by another venue. These retained daily closes are not live quotes, bid/ask spreads or offers to execute an order.

2. Apply a window before interpreting the result

The candle timestamp is the start of the UTC bucket [start, end), not its last trade or publication instant. Source capture is a separate clock. Unknown publication and freshness are shown as unknown, and a missing observation remains a gap.

3. Inspect one connected original day

Choose a recorded day with the date selector, previous/next controls, horizontal arrow keys or table row. One original identity updates the OHLCV card, chart marker and selected table row; switching candle, close, volume and table presentations retains it. Latest day explicitly returns to the newest retained row.

The selected candle Passport verifies its exact original price, BTC volume, UTC bucket and capture under current access. For a historical selection the summary capture is explicitly labelled as belonging to the latest result candle; the selected candle’s own capture is in its Passport. Chart coordinates are presentation only, while exact decimal strings remain available.

4. Preserve the complete original

Network estimates, sentiment, monthly returns and ETF flows are separate workflows with their own methodologies. Daily price history does not establish investment suitability, future returns or a trading recommendation.

1. Confirm your own units before using a source price

Load saved assignments or choose a current source for each own instrument and explicitly confirm what one recorded unit means. A matching symbol alone is not a verified broker or listing identity. The conversion panel can use these deliberate assignments directly; a separate valuation run is not required.

A portfolio containing only recorded cash can use the same conversion panel without a purchased asset or price assignment. Choose the economic cutoff, the exact CNB fixing day and CZK, EUR or USD, then Apply. The report uses your account balances and the two original fixing sources; it creates no holding. Cash remains signed, and a missing required conversion keeps the complete total unavailable. Once the portfolio has an open position, its price-source workflow applies again.

Asset prices remain original USD source values with their actual coordinates and captures. A daily Yahoo close is not a live quote, and its corporate-action basis remains explicit. An absent, too-old or unsuitable price leaves that position unavailable rather than substituting cost basis.

2. Choose a presentation currency and one exact fixing day

Choose CZK, EUR or USD and explicitly enter one supported2026 CNB fixing day within the source-input window. This changes only the report presentation. The portfolio accounting base, original transactions, cash balances, quantities and cost basis remain unchanged.

The exact selected fixing day is separate from each asset-price day and from source capture. A weekend or missing fixing stays absent; an earlier rate is never inserted. Source: CNB. The reference conversion assumes no executable trade, spread, fees, settlement rate or tax treatment.

3. Read coverage before interpreting a total

Recorded quantity times the exact original price gives the original-currency position value. The report then converts eligible positions and signed cash through the actual required CNB legs. Complete portfolio value is available only when every open position and cash balance has its necessary inputs; a known portion is labelled as such. Unsupported cash currencies are not treated as zero.

Finite monetary inputs and rates are combined as exact integer fractions. Rounding occurs only at each declared output boundary to18 decimal places, HALF_EVEN, with rounding flags. Totals use exact aggregate fractions before their final rounding, so displayed rounded rows are not silently reused as exact total inputs.

Allocation includes both holdings and cash in the chosen presentation currency. Holdings group by your recorded asset identity; cash groups by its original currency across accounts. With complete coverage the denominator is the whole reference portfolio value. Otherwise percentages describe only the known portion, and every group retains its covered/total count. This is a value allocation, not ETF look-through or economic currency exposure.

Allocation percentages use unrounded group values and the exact common denominator, then round independently to18 places HALF_EVEN. Rounded percentages may differ slightly from100 in total. Any known negative position or cash exposure disables ordinary allocation percentages and bars; signed values remain readable. A zero denominator has no percentage. JSON and CSV include these same groups, denominator, coverage, method and rounding flags.

4. Verify inputs and retain a coherent private report

One bounded source-input query and result manifest retain all selected price and FX originals. The private calculation manifest binds that exact result to your current ledger revision and economic inputs, assignments, chosen day and presentation currency. The price and FX Passports show original values, units, evidence and clocks; the report distinguishes derived amounts.

Full JSON/CSV reports recheck current owner, session, portfolio and source download access. Editing the draft or changing economic inputs retires the old report instead of letting a late response overwrite your selection. Calculations stay in page memory and are not written into the financial ledger or publicly shared caches.

My target allocation compares your own weights with this same whole valuation. Enter every instrument and cash target with an exact100% total, then explicitly review. Drift is current weight minus target in percentage points; gross adjustment is target amount minus current amount. Incomplete coverage, negative exposures and nonpositive whole values do not produce a whole-portfolio target model. At most50groups are supported. Costs, taxes, executable quantities and orders are not included. Separate Target model JSON/CSV exports retain the exact model and source identity. Unsaved drafts and calculated models remain temporary page memory. Your own weights have a separate reviewed Save to the private account and explicit Load into draft, available even before price valuation. Loading never runs a query or changes the ledger. If the saved groups no longer match the current portfolio, review a fresh complete target set rather than silently dropping a group.

5. Keep value separate from investment performance

A reference value using selected current source prices and a chosen fixing is not historical base-currency profit, TWR, XIRR or a tax result. Original-currency unrealized differences still require their own known cost basis. Deposits and withdrawals remain external cash flows, not investment gains.

6. Choose an economic valuation cutoff explicitly

Use the current recorded portfolio by default, or prepare an exact UTC cutoff and apply it explicitly. Applying reads a separate admitted projection of the currently recorded journal at that economic instant, with its own quantities, signed cash and reversal treatment. Changing a date does not relabel current holdings as historical. The fixing day, source coordinates, source capture and economic cutoff remain separate.

Then choose the supported fixing day and presentation currency and calculate explicitly with the existing source assignments. Prices use the latest admissible original source coordinates within the existing input window; choosing an old cutoff does not extend current source-age or coverage limits. An unavailable price or fixing remains unavailable, while the actual recorded holdings and cash stay readable. Yahoo daily Close coordinates do not establish that a close was executable or already published at an intraday cutoff.

Read the chosen cutoff alongside the complete input identity, coverage, position and cash values and allocation. Full JSON and CSV verify that same held cutoff again; export does not silently return to the current projection. Applying another cutoff or resetting to current inputs retires the previous report and late replies cannot replace the new choice. Current saved target weights and their editor remain separate from the dated report.

This is a retrospective reading of the currently recorded ledger using eligible original inputs, not a reconstruction of what you or the market knew then. It is not a historical wealth curve, TWR, XIRR, profit calculation, executable trade or tax conclusion. The date control writes no transaction, source price, fixing, target or accounting balance.

7. Include crypto while retaining original price and fixing evidence

The same valuation can use explicitly assigned original V4 crypto references beside supported equity sources. Search the actual captured asset list and check the exact CoinGecko ID and one-unit assertion. Old assignments remain their original source. At most eight selected price series are used; the full captured crypto cohort is retained separately as their original evidence.

One versioned input result binds the complete crypto capture, selected other prices and the two original CNB fixing series. Their source updates and capture clocks remain independent. The chosen fixing day is not the crypto capture time. An absent fixing is not carried from another day. A crypto price outside its one-hour application age limit or later than the economic cutoff cannot value that position; departed and null members never borrow an older price.

Read native price, converted amount and coverage separately. Partial known subtotals never stand in for complete wealth. Full valuation JSON and CSV include the same complete source artifact, selected unit relationships and conversion evidence, with current account and source access rechecked. Saving source assignments remains separate from saving a price snapshot or a calculated report.

1. Choose your current private list

Open watched equity prices while signed in, then choose one named list or the union of your lists. A valid empty list stays empty. A deleted or inaccessible named list is not replaced with All automatically. Changing the list does not create a trade or change membership.

The first Watchlist tab, Market overview, puts four separate price groups on one page. Choose a list once, then Apply selection and read prices. Global search and group filters change only this private presentation. Coverage keeps every selected member and complete counts. A source failure leaves its coverage unverified; it does not prove an absent or zero price. The original groups cover up to250 CoinGecko assets and eight Yahoo listings. The broader crypto catalogue group reads at most100 exact bound members per price page. More stock listings reads at most two exact bound listings per page, separately from the original eight-price source result.

The original crypto and equity groups retain their own canonical result, source clocks, Passport and whole export. The broader crypto price-page group retains its separate source input and whole applied-page export. More stock listings retains its own original quote inputs. Neither substitutes a canonical price result. Currencies are not combined or jointly ranked. Existing two-through-six comparisons stay within their own compatible family. An open source review, export or comparison locks shared selection and other groups. Membership changes require another explicit Apply; renaming or reordering preserves the same members. Manage lists stays available for names, order and membership.

In the full equity overview or a supported listing history, Follow adds that exact listing to your most recently updated list with room, or creates a default list if none is eligible. Following means membership in at least one owned list; removing it removes that listing from all your lists. Open Watchlist to choose names and ordering. The source-price result is unchanged by membership edits.

Private membership is checked against the current account and session. Source prices have their own independent access verification. If either proof becomes unavailable, private values and actions are concealed until verification succeeds; an old account response never replaces the current selection.

2. Match exact listings and read coverage

The original canonical equity group covers eight specific Yahoo listings: AAPL, MSFT, NVDA, AMD, GOOGL, AMZN, SPY and QQQ. Membership matches the exact catalogue listing identifier. A company, a similarly named instrument or an older source identity does not silently become that listing.

Compare watched item count, supported listing count and available source-price count before reading the table. An unsupported member remains explicitly without a price; it is not zero. Search, ordering and original-precision controls only project the same captured result. They do not fetch a replacement universe.

3. Check the original price and its own date

Each row retains the original Yahoo Close in USD per unit, its source coordinate and Datisto capture. The coordinate denotes the beginning of the source session interval, not a verified closing trade or publication instant. Rows need not have a common latest date.

Yahoo Close includes provider split adjustments; it is not Adj Close, total return, an ETF NAV or an executable venue quote. Open the price Passport to inspect the exact decimal, original evidence, revision, clocks and current source access. A detail link opens that specific listing history.

4. Keep the private filter separate from the original export

Changing your private list filters the same full source result rather than creating a different price calculation. The result identity remains visible. During a Passport or export review, list selection is locked so the open review does not drift to another membership context.

The original CSV/JSON export contains all eight source listings and their evidence, including listings hidden by your private filter. It does not contain your private list names or membership. Download access is verified again. This screen does not add a private watchlist Snapshot or grant redistribution rights.

5. Compare two through six watched listings

Below the price overview, explicitly select two through six currently qualified members. Nothing is preselected and unsupported identities stay separate. Review the exact listings and apply that selection to the comparison draft. Editing checkboxes alone leaves the applied comparison unchanged; a changed list or session cannot silently replace its membership.

Load source inputs starts the separate history request only for those selected listings. Apply interval then uses their captured original histories without another source read. Common completed timestamps drive price changes, observed drawdowns, graph and table. This is independent of the latest eight-price overview, not a reconstruction of its missing dates or portfolio performance. Comparison JSON/CSV includes the complete applied histories and manifest, without private list names, IDs or session details. No selection is placed in a URL or saved automatically.

6. Follow an exact broader-catalogue asset and read its price

In the full cryptocurrency catalogue, open the exact asset detail and choose Follow asset while signed in. This one explicit action verifies its source ID, then uses the existing owned-list command. It adds to an active owned list with room or creates a default list if capacity permits. Verifying an identity or choosing Verify my lists alone does not add membership. Equal symbols never identify the same asset.

Open Manage lists to find the stored asset and exact native ID. Open the exact asset in the full catalogue returns through an encoded ID link, preserving unusual characters without guessing a ticker or another coin. Unfollow removes the asset from all your owned lists. Known owned membership can still be verified and removed after the source catalogue expires; a new addition requires a current admitted catalogue. Resolve a retained uncertain command in the existing list recovery flow before another change.

Open prices in my lists, choose your current list or union, then explicitly Apply selection and read prices. Broader crypto catalogue · my selected prices reads the first applied page, up to100 exact IDs. For another page choose Price page, then Read this page. Changing the page choice alone leaves the prior applied page unchanged. Other pages are not verified by that read. An ID absent from the current catalogue remains visible instead of being replaced.

Read the separate catalogue and quote capture clocks, original SHA and provider last_updated. Price is the original USD per token provider aggregate, not an executable trade. Source null, an absent price field, an asset not returned and an ID absent from the catalogue are separate states; none becomes zero. Inspect the selected asset’s original fields and units. A changed account, session, membership or request hides obsolete private-context results.

Whole price page JSON and Whole price page CSV contain every original field and returned/not-returned state of the applied page, not prices from other watchlist pages. Current download access is checked again. If any applied member is absent from the current catalogue, the whole-page export is unavailable: no reduced file is produced. Following does not enable an alert, create a price observation or save price history; original history and its Snapshot remain separate workflows.

7. Follow and return to an exact broader stock listing

In Stocks → Catalogue, explicitly load and review a listing and its history, then choose Follow listing while signed in. The action verifies that exact original source and uses the existing owned-list command. Venue, instrument type, native currency and time zone identify the listing; a matching ticker or SEC candidate alone does not prove its share class. The original eight listings reuse their established identity only when the native listing matches.

Manage lists shows the stored symbol, venue, type and currency. Return to this listing opens /stocks/catalogue?watchEntity=… with the stored identity, not a guessed symbol search. Verify the owned listing, explicitly load its current SEC catalogue, then load and review the exact candidate’s listing and history. A missing, changed or ambiguous candidate is refused. The new native listing must match the saved venue, type, currency and time zone. Opening the return link or verifying owned metadata alone does not read Yahoo or replace the saved binding.

In Market overview, explicitly apply your list selection. More stock listings reads a page of at most two selected listings through the existing source reader. Other pages remain unread until explicitly requested. Each requested member keeps its own available or unavailable state; an unavailable source or changed listing never becomes zero or a substitute instrument. Watchlist membership, current source permissions and the current account/session are verified separately.

These are original provider regularMarketPrice values with their own regularMarketTime, native currency and capture clocks. They are not the completed daily Close of the original eight-listing group, an executable trade, a portfolio valuation or an FX conversion. A null price or unknown quote time remains explicit. Following does not create a price series, enable an alert or save history. Use the separate original-history workflow for its full files or Snapshot.

Unfollow removes the known listing from all your lists even after its source input expires. An uncertain membership command keeps its original recovery flow; verify its original receipt before another change. A new addition or price read needs its applicable current authority. No opening, return or page change silently follows a listing or places an order.

Once a complete price page has all original sources available, it automatically rechecks access roughly every 10 seconds against those same results, SHA hashes and capture times, without another Yahoo read. The first failed check stops automatic renewal; when its proof expires, the values are hidden. New prices require an explicit read.

Main workspace and exact source links

Opening DeFi and protocol TVL automatically reads the actual catalogue. Both inputs form one workspace while retaining each document’s own capture and hash. Switch protocols and chains, filter category or chain membership and inspect the exact original record. A protocol-ID or chain-name link never substitutes another record when it is absent from the source.

2. Start with the correct universe

Protocols and chains are two separate provider catalogues. The protocol response also includes CEX and other source categories. Keep the category visible and use its filter when you want a narrower group. A position in the table is an ordering of this response, not a recommendation or independent safety assessment.

3. Compare like-for-like TVL

TVL is the provider’s USD valuation of assets in the stated scope. It is distinct from token market capitalization, protocol revenue, yield and stablecoin supply. A change in asset prices can change USD TVL without a deposit or withdrawal. Reported percentage changes keep the provider’s own periods; they are not investment returns.

4. Read the detail without adding overlapping amounts

A protocol may list several chains and additional metrics such as staking or borrowed amounts. These labels do not form a guaranteed additive partition. Inspect the original labelled values without summing them into a new total. The separately supplied chain TVL has its own provider scope and need not equal a sum of visible protocol rows.

5. Keep clocks and missing values

Datisto shows when each original response was captured. If the source supplies no observation or update timestamp, the measurement time stays unknown. Capture time is not silently used as that timestamp, and two documents do not become a simultaneous observation. A missing value is not zero. Source categories and descriptions retain their original language.

6. Keep the entire selected result

Search, filters and sorting define the visible selection. Its complete CSV and JSON include the selected rows rather than only the current table page, together with source and capture context. The chart is a cross-sectional comparison of current values, not a historical curve. This local development catalogue does not create a saved account analysis or change a financial position.

Main catalogue and exact source detail

Opening Stablecoins and circulation automatically reads the actual catalogue. The table can show every asset or separate source-reported chain aggregates. Asset and chain links use the exact original ID or name; an absent record is never replaced by another. Filters, chart, detail and whole files retain the same captured batch.

2. Find a stablecoin and read its source status

Search the provider catalogue by name or symbol and filter by peg type, mechanism or chain. Switch the peg filter to all categories to inspect the whole catalogue. Source flags such as deprecated, delisted or deadFrom remain visible. An absent flag does not certify current activity or safety.

3. Separate price, peg type and circulating value

The provider quote uses USD per token. Circulating fields already contain USD valuations, including fields named peggedEUR, peggedJPY or peggedVAR. Those keys classify the peg; they do not turn the displayed amount into euros, yen or a native token count. Do not multiply these already valued amounts by the price again.

4. Treat earlier balances as revalued comparisons

The source values current and previous-day/week/month balances with the same current valuation price. These buckets are not a dated history of observed market capitalization. Exact underlying observation dates are absent from this response. Datisto retains the original bucket labels instead of assigning calendar dates or calculating investment returns.

5. Keep missing quotes and chain scopes explicit

A missing quote stays missing even when circulating value is present: the source can use an undisclosed peg FX fallback for valuation. Its rate and timestamp are not established by this payload. The separate chain aggregates exclude some double-counted assets and normalize chains, so do not reconstruct them by summing catalogue rows.

6. Inspect detail and keep the whole selected result

Open a row to inspect exact values, original field keys, source flags and capture context. Table and current cross-sectional chart use the same filtered selection. Complete CSV and JSON preserve all selected records, not only the visible page. The local preview does not create a saved account analysis, certify reserves or promise redemption at the peg.

1. Read area and colour separately

Last-known batch means the short freshness interval has elapsed. A transient source failure may retain a dated capture; an explicit source shutdown or withdrawn access conceals it. Refresh is explicit. Selection and export belong to the captured batch, not a historically reconstructed ranking for today.

A rectangle’s area follows its positive USD market capitalization within the current filtered batch. Colour follows the source price change for the chosen 1-hour, 24-hour, 7-day or 30-day period. A large rectangle is not necessarily a large percentage move. The legend identifies direction and the five-percent colour threshold.

2. Keep the loaded universe explicit

The map uses the existing CoinGecko batch of up to250 assets. Search changes the selection and therefore relative tile sizes. This is not global dominance or a historical membership reconstruction. A missing or zero market cap has no weighted area, but the asset remains in the source table and complete export.

3. Open a tile without losing the selection

Click a rectangle or use its keyboard focus to open the shared quick preview. Arrow controls move through the filtered order; closing returns to the map. Tiny rectangles may be smaller than a pixel, so every asset is also available through a table link and quick-preview button. A full-detail link opens the original asset screen.

4. Distinguish missing change, rounding and time

Zero and an unavailable change use the neutral colour, while the label distinguishes0% from a dash. Display coordinates and formatted amounts are approximate; original decimal strings remain in labels and exports. Each asset has its own provider update time. Batch capture time does not make every quote simultaneous or executable.

5. Retain the complete filtered data

The table, map and CSV/JSON use the same selection and source batch. Both exports include all selected rows, including assets with no map area, original values, query conditions and source fingerprint. Changing a colour period does not calculate a new historical return. The source stablecoin filter joins exact CoinGecko IDs only; its independent catalogue and source flags retain their own capture time and hash in exports. No match means classification is unestablished, rather than proof that an asset is not a stablecoin. Local watch/compare controls follow the existing preview semantics; an account analysis is not silently saved.

1. Choose the source change period

From the crypto market overview, open Gainers and losers. Last-known batch means the short freshness interval has elapsed. A transient source failure may retain a dated capture; an explicit source shutdown or withdrawn access conceals it. Refresh is explicit. Selection and export belong to the captured batch, not a historically reconstructed ranking for today.

The 1h, 24h, 7d and 30d periods refer to changes supplied by the provider. They are not a return calculated from two simultaneous quotes. Switching periods changes ranking and distribution over the same loaded assets; it does not retrieve historical market membership.

2. Distinguish the loaded universe from your selection

Summaries describe the current filtered result from a provider batch of up to 250 assets. Search or direction can narrow that result. The number of rising assets in this selection is not breadth across every cryptocurrency and is not a capitalization-weighted share. Stablecoin status is not inferred from a name or a price close to one. A separate filter uses only exact CoinGecko IDs stated in the DefiLlama catalogue. You can show source matches, USD or another peg, unmatched IDs or conflicting categories. No match does not prove an asset is not a stablecoin. The catalogue includes inactive records and has its own capture time; it verifies neither parity nor reserves.

3. Apply your own minimum market cap

The minimum market cap is an explicit USD amount. An empty condition means no such restriction, rather than an invented zero cap. Apply the entered minimum using the screen control. An invalid draft does not change the previous applied result. With an active condition, an unavailable market cap cannot establish that it meets the threshold.

4. Read the distribution and largest moves

A zero change and an unavailable change have distinct meanings. Distribution uses the stated fixed boundaries; a nearby value is not classified from its rounded label. Gainers and losers are ranked by original decimal values. Short lists show the largest moves, while the table and export retain the complete result.

5. Open the underlying asset and retain context

Quick preview and full detail use the exact original asset ID. A comparison selection of two to six identities stays only in this tab’s memory, rather than your account. Seven-day points without timestamps are not a dated history; the full detail reads a separate dated history automatically. Each quote has its own provider timestamp; capturing the batch does not make every price simultaneous. CSV and JSON contain the full filtered result, conditions, original values and source fingerprint. This local work does not itself create a saved account analysis or a trading instruction.

1. Distinguish global coverage from the selected 250 assets

From the main crypto market overview, open Global indicators. The full source view uses a separate CoinGecko /global response. Its market cap, volume and shares are not a sum of our 250-asset table. Coverage follows the assets and exchanges tracked by the provider; it does not establish complete coverage of every asset or trade worldwide. Counts are original source statistics, rather than the number of items on this page.

2. Read market cap and volume in their own units

The main cards retain original market cap and rolling 24-hour volume in USD. Complete source currency maps have separate rows with original keys and availability; they are not converted with a custom exchange rate. They measure different things and are not added together. The reported volume is not repaired from Bitcoin prices or our exchange tables. The source 24-hour volume change is not a price return. Cards may round values; the exact string remains in the source view and exports.

3. Shares are original percentages, rather than a new pie

BTC, ETH and other keys have provider-supplied market-cap percentages. The chart retains each percentage, fills no remainder to 100% and invents no Others category. Keys are labels in this source field; they are not automatically verified token or trading-pair identities. No change in share is inferred from an earlier capture.

4. Distinguish provider time, capture time and missing data

The provider updated_at is converted from its original Unix seconds to UTC. Response capture is a separate time; refreshing the page does not replace the provider timestamp. An absent field, explicit null and supplied zero remain distinct. A failed refresh may retain the previous capture with its original time. A single overview does not create a historical curve.

1. Open a specific asset

Open a full cryptocurrency detail from Markets, the map or the movers overview. Dated history uses its exact CoinGecko ID from the bounded market selection. A symbol shared by two different assets is no reason to combine their data and is not used to infer an exchange pair.

2. Choose a history period

The public detail automatically reads seven-day history for the exact identity; changing to 30 or 90 days reads that range. This is a relative provider request, not a guarantee of complete or regularly sampled points. A new or unavailable asset may have shorter history. After a completed failure, retry with the control; the application does not loop retries.

Original seven-day points from the market batch are a separate quick input without exact timestamps. Adding the new history does not turn their order into invented hourly or daily dates.

3. Explore actual points in UTC

Chart and table select the same original point. Horizontal position follows its actual time, rather than evenly spaced row order. Details retain original Unix milliseconds and UTC. Response capture is not the time of every measurement. A price point is not automatically a daily closing price or exchange OHLC.

4. Distinguish price, market cap and volume

Price is USD per token, market cap is USD, and volume is the original rolling 24-hour USD measure. Each series retains its own timestamps; tabs do not force them onto common dates. Missing or null is not zero, is not filled with a previous price and creates no new observation. Chart coordinates are approximate; original numeric strings remain exact.

5. Keep the whole result with its sources

CSV and JSON retain every original series and point, even when one tab or table page is visible. They include exact identity, requested period, separate history capture and fingerprint, and the market-selection receipt that established identity. The two capture times remain distinct. This local history is not automatically saved as a canonical account analysis.

1. Choose a fund and review the input

Default CSPX loads automatically; changing to IUSA reads its own holdings. The result directly shows the CSV date, total row count and capture time. Refresh data repeats the read. A transient failure retains the dated result; withdrawn access conceals it.

2. Search the complete table

Search name or ticker, select a reported asset class, and choose source order, descending weight or name. Changes first form a draft; Apply filters changes the result. All rows are retained, including cash, collateral and futures when reported. The equity-holding count on a product page therefore need not equal the file row count. Paging by 25, 50 or 100 rows does not change coverage.

3. Inspect one row

Click a name to open one inspector containing every original field, order and string. IUSA may additionally have Type. Weight is labelled %. Market Currency is separate and does not establish the documented currency or unit of Market Value, Notional Value or Price. No conversions, totals or derived prices are calculated. Cash prices may use a different scale; do not treat them as a one-unit price or exchange quote.

4. Distinguish dates and provenance

The holdings date comes from the CSV’s own preamble. Capture and cache expiry describe the downloaded document and do not replace that date. Row provenance shows its logical position, product ID and file SHA-256. The ISIN association comes from the exact product address; the CSV itself does not assert the fund ISIN. This view provides neither holdings history nor every catalogue product’s holdings.

5. Download the whole same result

Complete JSON or CSV retains the applied query, original source, every fund row and filter matches/exclusions. The open detail and page do not limit exports. Download first verifies the same retained input and its hash without refreshing the issuer. If the local server no longer retains it, explicitly load a new input. This is not an automatically saved account analysis or canonical historical Snapshot.

6. Open sectors, locations and asset classes

Allocations use every original row of the applied fund, independently of table search and filters. Switch Sectors, Locations or Asset classes. Each group has the exact sum of known reported weights and counts of known, missing, zero and negative weights. An empty source classification or dash stays Not reported in source; the original text remains in row detail.

Signed bars share a zero baseline. Reported rounded weights are not rescaled to 100%; their sum may exceed 100%. Click a group to open all members, paged by 25, and then an original row. An unknown weight is not zero. Original Location is not a revenue or geographic-risk allocation. Money amounts, prices and currency risk are neither aggregated nor inferred.

Complete JSON/CSV contain all three allocations, weight coverage and every member reference alongside original rows and one source document. The open group and table filters never change those aggregates. The allocation date is the CSV’s own date, not the new computation time.

1. Prepare an applied selection

Load and review the original CSPX or IUSA CSV, then explicitly apply it. Apply your search, asset-class and order filters as well. Choose Sectors, Locations or Asset classes, optionally an exact group, and 25, 50 or 100 rows per page. Saved settings use applied filters and the current presentation, rather than unapplied filter edits.

2. Review and confirm a name

Sign in, name the selection and choose Review save. Review the fund, filters, order, page size and allocation group before confirming. Only Confirm save to account sends the command. Cancelling before sending stores nothing. Names and settings are private.

3. Find and review the saved settings

Find Saved fund holdings settings in My analyses, with a compact list in Return to work. Open the exact saved link. Fresh account verification shows the configuration without downloading CSV, calculating or applying a result. Adopting the reviewed conditions fills only a draft. If work is already in progress, explicitly decide whether to discard it first. Review a changed saved version again; a removed entry cannot be adopted, and expired verification requires account re-verification first. Renewing verification of the same version preserves a valid review.

4. Load a new input

After adoption, explicitly load the issuer input, review the CSV’s own date and apply it. The exact saved allocation dimension and group are restored. If the group is absent from this file, an unavailable selection appears without substitution. Opening account settings does not promise the same holdings or date as when saved.

5. Rename, remove or resolve an uncertain save

In My analyses, rename or remove settings through review and confirmation. An error proves neither success nor cancellation. If the command might have been accepted, explicitly retry the same retained request; it is never automatically sent and its key and content stay unchanged. Changing accounts conceals private information.

6. Distinguish settings from a financial export

The account stores only a name, query and presentation. It does not store issuer CSV, positions, weights, values, original clocks or a financial Snapshot. Complete holdings JSON/CSV is separately downloaded from an applied result, including all original rows and allocations. Settings alone are not a ValuePassport or a publicly shared analysis.

1. Choose the exact fund and load the input

Default CSPX NAV loads automatically; switching to IUSA reads its own original series. The initial period uses actual coverage. Check the fund, date and point range. Refresh data repeats the read without filling missing points.

2. Apply a window

Prepare a date window and explicitly apply it. Draft edits never relabel the previous chart, table or export. The chart and complete table use one query and the same result. Choose a shorter window or the complete available range without silently selecting only some points.

3. Read actual NAV points in USD

The line includes every original point in the applied window. Missing calendar dates are not inserted, unavailable points are never replaced with a previous value and the line breaks at an unavailable point. This is not an intraday price. Benchmark, priceData and performance series are never substituted for NAV. Do not infer dividend, split or adjusted conventions from this view.

4. Inspect an exact point and its provenance

Select an original position in the table or point control. The inspector retains the exact numeric lexeme and source formatting, the point’s own date and original date field. A point date is not a publication time. HTML capture, original document hash and extracted fragment hashes are separate evidence; capture never rewrites old point dates.

5. Download the whole same result

Complete JSON/CSV retains the applied query, manifest, original source fragments, all points and explicitly retained points outside the window. Pagination and selected point do not narrow the export. Download verifies the same retained input without refreshing the issuer. If the local server no longer retains it, explicitly load a new input. This is not a canonical account Snapshot, sharing or a ValuePassport.

1. Choose a fund and original input

Default CSPX growth and benchmark load automatically; changing to IUSA reads its own input. The initial interval uses this growth series’ coverage, not the NAV range. Check the source title, named benchmark, original methodology and dates.

2. Distinguish growth from NAV

Growth and benchmark are distinct original performanceData and benchmarkData series. NAV in USD has its own screen, method and latest point date; switch views for the same fund. NAV is never overlaid on the hypothetical-growth axis, and priceData or performance percentages are never substituted.

3. Apply a window without rebasing

Choose and explicitly apply a window. Chart and table retain original issuer levels and the USD 10,000 starting base even when a later period is selected. No return, alpha, tracking error or personal portfolio benchmark is calculated. Original points outside the window remain in complete exports.

4. Retain asymmetric dates and gaps

The table uses the union of actual dates from both series. A benchmark-only date remains visible without a substituted fund value. Each line breaks at its own absent or null point. There is no calendar filling, carry-forward, smoothing or silent sampling.

5. Read the exact original methodology

The selected title and disclaimer of this particular page are shown as plain text. This evidence describes NAV basis and reinvestment of gross income where applicable. A different catalogue disclaimer about net income is not carried over. The exact price/net/gross/tax index variant is unestablished; tax-equivalent comparison is not promised. This is an original hypothetical illustration, not an exchange quote or guaranteed outcome.

6. Inspect paired points and download the whole result

Select a date in the table or point control. The inspector retains the exact value lexeme, formatting, own date and original position separately for each series. A source date is not publication or trade time. Complete JSON/CSV retain the same query and manifest, both original series, union dates, missing counterparts and methodology evidence. Selected point and page do not limit exports. Download verifies the same retained input without refreshing the issuer; this is not a canonical account Snapshot, ValuePassport or public share.

Select a fund and view

Choose a fund and Facts and listings, NAV, Growth and benchmark or Holdings. The first visit and changes of fund or view load and show the corresponding actual input. Source dates, including the holdings CSV’s own date, remain visible. Apply edited local filters and custom dates with their control; an unfinished draft is not labelled as the applied result.

Opening Holdings also reads the selected fund’s original CSV. Other views do not need it. Refresh data repeats the read; the issuer is not continuously polled. A transient error may retain the dated result, while explicit access withdrawal conceals it. Cache expiry governs application reuse, not market currentness.

Original identities and facts

Facts preserve each field’s own value, formatting and date; listings retain every original column. Select an item to open its single provenance inspector. Catalogue ticker CSX5 and native ticker CS51 remain distinct within the exact selected product. They do not establish current venue identity.

NAV and original growth

Both historical windows include their endpoints. The table retains every original point. An absent counterpart or null value is never filled with zero or a preceding observation; the affected line breaks. NAV is neither exchange price nor total return.

Original hypothetical growth retains its 10,000 base and issuer methodology; selecting a window never rebases it. AGGH has an EUR fund and USD benchmark: each has its own panel and axis, with no currency conversion or relative-performance claim. The benchmark’s exact tax or return variant remains unknown.

Whole original holdings

The table shows every loaded position and the CSV’s original date. Search is applied explicitly; pagination changes presentation only. Details retain every original column, including bond fields. Cash, derivatives and duplicate tickers are never merged. Market Currency does not establish the unit or currency of Price, Market Value or Notional Value; no money totals or FX conversion are calculated.

The same whole input in a file

Complete JSON/CSV retain the same query, result, manifest, original fragments and every point and row including exclusions. A loaded CSV remains complete. Before export, the current local gate verifies the exact same retained identity and hashes of fragments, selected evidence and CSV. Export never refreshes the source or substitutes newer values.

Save the applied selection

On six-fund data, select a fund and view, load and review an input, then apply it. Enter a name under Save applied settings and choose Review save. Review shows fund, view, whole or inclusive custom window, holdings search and current page size. Only Confirm save to account sends the change.

The applied query is saved, not an unfinished filter draft. Page size is the current presentation choice. NAV points, hypothetical growth, holdings, source files, source identifiers and capture clocks are never stored in this configuration. Result JSON/CSV is a separate tool that retains the complete original input.

My analyses and Return to work

My analyses lists your named six-fund selections. Rename and Remove prepare a review and require explicit confirmation. A compact list also appears in Return to work on the dashboard. Each link opens the exact private settings ID; another user cannot adopt it as their own.

Private names and conditions are shown only after fresh account verification. A missing or removed entry is never substituted with another fund. Saving settings does not publish them or create a canonical Snapshot.

Review and adopt into a draft only

Opening freshly verifies your collection and displays the original selection readably. Adopt reviewed conditions into draft fetches and calculates nothing. If you already have an active draft or result, explicitly decide to discard it first. If saved conditions change during review, prepare the current saved version for review.

Then separately choose Load input for review, inspect its own dates and apply it. Holdings CSV is read only when explicitly loading Holdings. Source age is assessed against the newly loaded input; saving settings does not extend cache life. AGGH remains an EUR fund and USD benchmark, with separate axes and no currency conversion.

Uncertain outcomes and retained requests

An uncertain outcome retains the exact request. Verify and retry the same request uses the same key and conditions; reopening sends nothing automatically. Leaving the page does not prove cancellation of an already sent change. After a confirmed conflict, verify the collection and explicitly prepare a new draft. An older receipt never replaces the current newer collection.

A reference is not an identified contract

Opening Commodities automatically reads the bounded actual reference overview. The fixed cohort is GC=F, SI=F, HG=F, CL=F, NG=F, ZC=F, ZW=F and ZS=F. These are provider futures references, rather than spot prices. Actual tradable contract identity, expiry, settlement and roll methodology are not established or inferred from a label. Unavailable symbols stay listed; another commodity is not substituted for a missing selection.

Each reference retains its own unit

Gold and silver retain USD per troy ounce, copper USD per pound, WTI and Brent oil USD per barrel and Henry Hub gas USD per MMBtu. Corn, wheat and soybeans retain source currency USX and unit USc/bushel: US cents per bushel, rather than USD. The application neither converts nor sums or price-ranks these unlike units. Original negative prices, zero and null remain distinct. Reported percentage change is not an independently calculated daily return; its comparison baseline was not supplied.

An explicitly applied window and whole files

Filter the table by name, symbol, unit and category. Select an exact reference and confirm its dates with Apply window. Chart, exact table and JSON/CSV use the same whole result of selected completed provider points, up to 120 days. Timestamps identify native provider daily buckets, rather than established settlement. Gaps remain unfilled. Table pages do not narrow export; files retain native units, negative values, zero, null, inputs and manifest.