Finance data, read the way an analyst reads it
Filings, statements, market and transaction records for models that have to get the numbers right: units and periods normalised, restated figures kept apart, entities resolved across identifiers, and account data masked.
Part of Runix Data. This page sets out the rules we apply in this domain and the public standards they follow.
filings XBRL: US GAAP and IFRS taxonomies entities LEI (ISO 17442), SEC CIK instruments ISIN (ISO 6166), FIGI venues MIC (ISO 10383) currency ISO 4217, with scale and sign periods fiscal period, as filed
A reference card. Each line names a public standard or convention; the rules below say how it is applied.
02 · Covers
The data this covers
Cleaned and structured from what you provide or have the rights to use, or built to a specification agreed in writing.
Regulatory filings
Annual and quarterly reports and their amendments, with the tagged financial facts inside them.
Financial statements
Income statements, balance sheets and cash-flow statements, figure by figure, with the period each belongs to.
Market and reference data
Instruments, venues and prices, tied to the identifiers the market already uses.
Transaction records
Ledgers and payment records, with account data masked before anything else happens.
03 · Rules
The rules, and where they come from
Each rule follows a public standard or an established practice in the field, named with it, so you can check the reasoning rather than take ours on trust.
The same unit before any comparison
Scale, currency and sign are normalised before two figures are compared: a value reported in thousands is not a value reported in units, and a cost shown as positive in one filing is negative in another.
Follows Numeric XBRL facts carry a unit and a stated rounding (the decimals attribute); ISO 4217 currency codes.
Periods as filed
Every figure keeps the reporting period it belongs to. Many fiscal years do not end on 31 December, and a balance at a date is not a flow over a quarter.
Follows XBRL's instant and duration contexts.
Restatements kept apart
A restated figure replaces an earlier one, either in an amended filing or as a revised comparative in a later report. Both are kept, with which supersedes which, so a corrected number is never presented as current.
Follows Amendment forms such as 10-K/A and 10-Q/A in SEC EDGAR, and revised prior-period comparatives in later filings.
One entity, many identifiers
Tickers change, companies rename and merge. Each record is tied to a stable identifier, the LEI where one exists and the SEC's CIK for anyone that files with the SEC.
Follows The Legal Entity Identifier (ISO 17442) maintained through GLEIF; OpenFIGI for instruments.
Frameworks not mixed
US GAAP and IFRS figures are kept apart, and are only combined through a mapping you can inspect.
Follows The separate US GAAP and IFRS XBRL taxonomies.
Account data masked
Account and card numbers and personal data are masked before a record reaches a training set. A record that cannot be cleared is held back, not passed through.
Follows The long-standing convention of showing only the last digits of an account or card number.
04 · References
Public references
The standards and open sources these rules are built on. They are other organisations' work, linked so you can read them yourself.
- XBRL InternationalThe not-for-profit consortium behind XBRL, the open standard for tagged business reporting.
- SEC EDGARUS public company filings, including their XBRL data.
- IFRS Accounting TaxonomyThe XBRL taxonomy for IFRS financial statements.
- GLEIFThe foundation that runs the Global LEI System and publishes LEI data.
- OpenFIGIOpen identifiers for financial instruments.
- ISO 4217 currency codesThe current code list, from SIX, the standard's maintenance agency.
05 · Delivery
What every delivery carries
The same in every domain; the Runix Data page has the full list.
Provenance and licence, per record
Where each record came from, what was done to it, and the licence or permission it was used under.
A quality report
Coverage, duplication and the checks each record passed, plus what was dropped and why.
Evaluation kept apart
Evaluation data split from training data by source, so a score is not inflated by near-duplicates.
06 · Questions
Common questions
Which accounting frameworks?
US GAAP and IFRS, kept apart. Figures are only combined across frameworks through a mapping you can inspect.
Can a model learn a number that was later restated?
Not as the current figure. Restated figures are linked to the ones they replace, whether the correction came in an amendment or a later filing, and the delivery says which is current.
Do you sell ready-made Finance datasets?
Not off the shelf. Runix Data builds to a specification agreed in writing: from data you provide or have the rights to use, or from public sources whose licences permit your use. The rules on this page apply either way.
What happens to the data we send?
It is processed only to do the work you asked for. It is not used to train models, ours or anyone else's, and it is not sold.
Send us a sample of your financial data
A slice of the real data and what the model has to do with it. We reply within one business day, and the scoped plan that follows includes the parts we think are not worth doing.
Request early access