Fundamental Data

Fundamental data describes the underlying businesses and economies behind market prices: revenue, earnings, cash flow, balance sheets, analyst estimates, ownership, and macroeconomic statistics. Discretionary investors read it company by company; quantitative traders process it for thousands of companies at once to build value, quality and growth signals. Using fundamental data correctly requires care, because when data was available, and how it has been revised, can make the difference between a realistic backtest and a fantasy.

Types of fundamental data#

Where it comes from#

• Company filings: 10 K, 10 Q and 8 K reports on the SEC's EDGAR system, with structured XBRL data.
• Commercial vendors: firms such as S&P Global (Compustat, Capital IQ), FactSet, Bloomberg, LSEG and Morningstar standardise data across companies.
• Free sources: EDGAR, company websites and some free APIs, with more cleaning required.
• Government agencies: for economic data, such as the BLS, BEA and Federal Reserve (FRED database).

Point in time vs restated data#

Companies revise past figures, and vendors may update their databases with restated numbers. A backtest that uses today's version of historical data may use numbers that were not available at the time.

Reporting lags#

Fundamental data becomes available only when it is released. Quarterly results arrive weeks after the quarter ends; annual reports even later. A common backtest rule is to assume data is available only after a lag, such as 45 to 90 days after the period end, unless precise filing dates are known.

Standardisation challenges#

• Different accounting standards: US GAAP vs IFRS.
• Different fiscal years: must be aligned to calendar periods.
• Adjusted vs reported figures: non GAAP numbers differ by company. See Earnings Quality and Cash Conversion.
• Currency conversion for international companies.
• Industry differences: banks and insurers have very different statements.

How quants use fundamental data#

See Factor Investing Explained.

Common pitfalls#

• Look ahead bias from restated or late data.
• Survivorship bias from missing delisted companies. See Survivorship and Selection Bias.
• Mismatched dates between prices and fundamentals.
• Outliers from tiny denominators, such as P/E with near zero earnings. See Outliers and Robust Statistics.

Frequently asked questions#

What is fundamental data?#

Information about the financial health and performance of companies and economies, such as earnings, balance sheets, analyst estimates and economic statistics.

What is point in time data?#

Data stored exactly as it was known on each historical date, including original values before any restatements, to avoid look ahead bias.

Where can I get fundamental data?#

From company filings on the SEC's EDGAR system, commercial vendors such as FactSet and S&P Global, and government statistics agencies.

Next, learn about news as data in News Feeds.