For researchers

Why the filing date is the product

A company's 2019 revenue is not one number. There is the figure it first published in 2020, and there is whatever the figure became after a restatement, a reclassification, or an auditor. A database that stores one value per company-year has quietly chosen the second, and has thrown away the first.

For most purposes that is fine. For anything where the question is what was knowable at a moment — an event study, a credit model, a trading rule, a policy evaluation dated to an announcement — it is not, because the model gets handed a number nobody could have had. The effect has a direction: restatements are not random, and neither is which firms survive to be restated.

Here, the first published figure stays. A later filing is added beside it, never over it, and a company that is struck off keeps everything it ever filed. Ask for a company as_of=2019-03-01 and you get exactly the rows whose filings were public on that date — not a filter applied afterwards, but the only rows that exist for that date.

If your current source overwrites on restatement, or drops firms when they exit, no amount of care downstream reconstructs what it discarded.

What is in the file

Counts are read from the delivery manifests, not typed into this page, so what is quoted here is what arrives. Current vintage 2026-09-28.

Delivered as gzipped CSV, one file per market, UTF-8 with a header row, plus a manifest carrying the row count and a SHA-256 for each file. Columns are documented in the data dictionary. Every row states its reporting basis, so a group consolidation is never silently mixed with the parent's own accounts.

What it is not

You would find all of this yourself in the first hour. Better that it is here.

• This is filed accounts, not a firm-level panel with employment, ownership or trade linkages. There are no director records, no shareholder chains and no group structures.
• Coverage begins where each register's structured filings begin, not where the companies do: Belgium from April 2022, France substantially complete from 2017, Denmark from 2013, Japan from 2016. A company older than its register's structured era has no rows for those years here.
• Turnover is not universal. Small-company reporting exemptions mean many filers publish a balance sheet and no revenue line; it is left empty rather than imputed. Measured fill rates per metric and per market are published on the methodology page — read them before you design around a field.
• Figures are reported as the filer published them, including their own rounding. Some Japanese filers present the five-year summary in millions of yen; that is their presentation, not our truncation.
• Names and identifiers come from the registers. Cross-border entity matching, LEI linkage and parent-subsidiary resolution are not included and would have to be done by you.

Price, and how an institution pays for it

Larger scopes: a single market with redistribution rights, all markets at US$7,500, or redistribution across everything, are set out on the pricing page, and the terms are in the , published in full before you buy rather than after.

Citing it

Cite the vintage, not just the dataset — it is the vintage that makes a result reproducible, and two vintages are not the same file.

WU Capital Limited (2026). Trading Agent point-in-time company fundamentals [Data set]. Vintage 2026-09-28. Auckland, New Zealand. https://www.tradingagentapp.com/fundamentals

@misc{tradingagent_pit_fundamentals,
  author       = {{WU Capital Limited}},
  title        = {Trading Agent point-in-time company fundamentals},
  year         = {2026},
  note         = {Data set. Vintage 2026-09-28},
  howpublished = {\url{https://www.tradingagentapp.com/fundamentals}}
}

Source registers must be credited in their own words as well; the required strings ship with every API response and bulk delivery, and are listed on the methodology page. For Japan in particular, attribution is a condition of the licence rather than a courtesy.