Choose the stock data your model actually needs: historical OHLCV, a security master, SEC filings and documents, point-in-time company facts, earnings events, source-linked news, macro observations, and dataset coverage. DataCedar exposes these streams through one versioned API so a buyer can test the schema, rights, history, and missing-data behavior before committing engineering time.
A stock market dataset is more than a table of prices. A research-ready dataset identifies the security behind each symbol, states which sessions and records should exist, separates effective time from when information became knowable, preserves source lineage, and reports whether absent data is empty, delayed, partial, restricted, or missing.
Query source-labeled OHLCV by symbol, interval, and elapsed date range. Keep raw and adjusted views distinguishable, retain the market session key, and inspect expected-versus-actual coverage before using the series in a backtest, chart, feature store, or model.
Resolve a current or historical ticker to a canonical security record before joining datasets. Issuer identity, listing attributes, ticker changes, and delisted states prevent the current-symbol shortcut from corrupting a historical universe.
Join filing chronology, accession metadata, primary documents, exhibits, and SEC/XBRL company facts to the same security identity. Accepted timestamps and known-at filters preserve the information set available at a historical decision time.
Add earnings dates, company events, public news references, and macro observations without flattening every timestamp into one date. Citations, revisions, source rights, and coverage state remain inspectable beside the normalized record.
Product boundary: DataCedar is not a consolidated real-time exchange feed and does not promise unrestricted redistribution of every upstream field. Available history, latency, text access, and export rights depend on the dataset and plan. Coverage responses expose those limits so buyers can reject an unsuitable stream before production use.
curl "https://api.datacedar.com/v3/stocks/assets?q=AAPL&status=active&limit=10" \
-H "X-API-Key: $DATACEDAR_API_KEY"{
"data": [{
"symbol": "AAPL",
"name": "Apple Inc.",
"asset_class": "stock",
"status": "active",
"source": "sec_nasdaq_security_master"
}],
"coverage": {
"status": "available",
"checked_at": "2026-07-31T00:00:00Z"
},
"pagination": { "next_cursor": null }
}Examples document the public contract and may use illustrative values or redacted identifiers. Availability fields and rights filters are authoritative for the active environment.
Write down the target universe, history window, bar interval, adjustment policy, evidence streams, known-at rule, and acceptable missing-data threshold.
Test the security catalog and coverage endpoints first. Confirm available symbols, earliest eligible history, latency, rights, and partial states before downloading rows.
Use canonical security identity and explicit timestamps to connect bars with filings, facts, earnings, news references, events, and macro observations.
Save the query, cutoff, cursor, schema version, coverage response, and source identifiers beside the model artifact so the run can be investigated and repeated.
A large dataset can still be unusable if it omits delisted securities, rewrites historical fundamentals with the latest restatement, or cannot explain a missing session. Define the decision the data must support and the historical information boundary before comparing vendor totals.
For price research, specify the universe, exchange calendar, interval, adjustment view, and corporate-action treatment. For event research, specify which filings, earnings records, news references, or macro series must be observable by a given known-at cutoff. These requirements turn a vague data purchase into a testable acceptance plan.
Ticker strings are not permanent identities. Mergers, relistings, share classes, symbol reuse, and delistings all create false joins when a dataset keys history only by the symbol visible today. A security master should resolve the issuer and listing represented by every market row.
Time needs the same discipline. An earnings period, filing acceptance, publication time, retrieval time, and dataset known-at time answer different questions. Coverage and rights are also data: an empty array caused by an unavailable source must not look like a verified absence, and a retrievable field must not be assumed redistributable.
Test the provider with ordinary symbols, thinly traded names, a ticker change, a delisted case, a split window, a filing amendment, and a known period with market holidays. Compare expected records with returned records and require the API to explain material differences.
Then run the exact downstream path: pagination, retries, incremental updates, a frozen historical query, and an export into the storage or model format you will operate. Measure engineering effort as well as subscription price. A cheap feed that needs a second identity vendor and extensive gap repair can be the expensive choice.
Price is only one line in the decision. The durable choice is the dataset that makes its historical and operational boundaries testable.
DataCedar exposes stock OHLCV, a security master, SEC filings and permitted documents, point-in-time SEC/XBRL facts, earnings events, company-event history, source-linked news metadata, macro observations, and coverage records through one API.
Build historical stock price and volume exports with explicit adjustments, coverage checks, symbol history, and event windows for reproducible backtesting.
Query permitted stock price and volume beside SEC filings, earnings events, point-in-time fundamentals, news references, macro data, and coverage metadata.
Query SEC filing metadata, primary documents, exhibits, XBRL facts, and amendments with accession-level provenance and known-at timestamps.
Query scheduled, confirmed, rescheduled, and completed earnings events with known-at history, source evidence, fiscal periods, and reaction windows.
Create a free key, inspect the universe and coverage state, and run representative stock, filing, fundamental, earnings, and news queries before you buy.
Start free