Datasets
Entity resolution API (LEI ↔ CIK ↔ CNPJ ↔ ISIN ↔ ticker)
Every financial dataset uses a different name for the same company, and the mapping between them is where integration projects go to die. A US filing knows a CIK, a European report knows an LEI, a Brazilian registry knows a CNPJ, a trading system knows a ticker that was reassigned to someone else four years ago. Joining them by name is how a subsidiary gets confused with its parent.
Hindcast treats identity as infrastructure rather than a lookup table. Identifiers resolve to entity nodes, relationships between parents, subsidiaries and issuing vehicles are explicit, and the mapping is versioned — because a ticker that pointed at one company in 2016 may point at another today, and a query about 2016 should get the 2016 answer.
Every other dataset in the catalog is joined through this graph. That is what makes a question spanning a 13F, a CVM portfolio and a price series a single query rather than three exports and a spreadsheet.
hindcast — session
previewhindcast> resolve a CNPJ to its ticker and LEI
→ tool: entity_resolve { id:"00.000.000/0001-00", type:"cnpj" }
✓ entity node with linked ticker, ISIN, LEI and CIK where they exist
source: registry records, mappings versioned by effective date
What's covered
- LEI, CIK, CNPJ, ISIN, CUSIP and exchange tickers mapped to one node
- Parent, subsidiary and issuing-vehicle relationships made explicit
- Versioned mappings — a reassigned ticker resolves correctly by date
- Name history so a renamed company stays findable under its old name
- Cross-jurisdiction joins between US and Brazilian records
- Every other catalog dataset keyed to the same entity identifiers
Join the waitlist — this page's dataset is already selected.
Questions
What happens when a ticker is reused by a different company?
The mapping is dated. Resolving that ticker as of an old date returns the company that held it then, which is the only behavior that keeps historical joins from silently attaching the wrong price series to the wrong filings.
Are subsidiaries rolled up to the parent automatically?
No — the relationship is exposed and you choose. Automatic roll-up is a modeling decision that belongs to your analysis, not to the data layer, and hiding it would make some exposure questions unanswerable.
Related
Join the waitlist. Vote the roadmap. First in, first served.