hindcast

Use cases & endpoints

Point-in-time fundamentals API — as-reported financials

A fundamentals database that stores one value per metric per period is answering the wrong question. Companies restate, auditors force adjustments, and accounting standards change — so the earnings figure for a quarter five years ago may have three different values depending on when you asked.

A point-in-time fundamentals layer keeps all of them. Each value carries the filing that produced it and the date that filing was made public, which turns an ambiguous question into a precise one: not what were earnings in that quarter, but what were earnings believed to be on this date. The second question is the only one a backtest can legitimately ask.

Hindcast builds fundamentals this way from the filings themselves rather than from a vendor snapshot. Restatements arrive as additional observations against the original, and any query with an as-of date resolves to the value that was current then.

hindcast — session

preview

hindcast> earnings per share for a past quarter, as first reported

→ tool: fundamentals { ticker:"...", metric:"eps", period:"...", as_of:"..." }

the originally filed value with its filing date

source: SEC filing that first reported the figure

Preview of the shape of an answer — endpoints open to the waitlist first.

What's covered

Join the waitlist — this page's dataset is already selected.

Need this one?votes for SEC & EDGAR
Optional — helps us build in the right order. All skippable.

Role

Company

Datasets you need

Questions

Isn't the restated number the more accurate one?

For understanding the business, often yes. For evaluating a decision made in the past, no — the market priced the original figure, and a model that trades on numbers published later is measuring foresight it did not have.

How far back does the revision history go?

As far as the filings do. The history is reconstructed from the source documents rather than from a vendor's current snapshot, which is what makes older vintages recoverable at all.

Join the waitlist. Vote the roadmap. First in, first served.

← Back to the data catalog