Fama-French Analytics

Data Sources

The reliability of any financial analysis depends on the quality of its underlying data. This dashboard relies exclusively on high-quality academic data sources that have been the foundation of decades of peer-reviewed research in asset pricing.

Primary Source: Kenneth R. French Data Library

All factor return data, portfolio returns, and characteristic sorts are sourced directly from the Kenneth R. French Data Library at the Tuck School of Business, Dartmouth College. This library is maintained by Professor Kenneth R. French and is widely regarded as the gold standard for academic finance research.

Underlying Data

The French Data Library constructs its portfolios and factors using data from two primary sources:

  • CRSP (Center for Research in Security Prices): Provides stock prices, returns, shares outstanding, and market capitalization data for all NYSE, AMEX, and NASDAQ listed stocks. CRSP is maintained by the University of Chicago Booth School of Business.
  • Compustat: Provides accounting and financial statement data, including book value of equity, earnings, cash flow, and dividends used to calculate valuation ratios such as Book-to-Market, Earnings-to-Price, and Cash Flow-to-Price.

Datasets Used

  • Fama-French 3 Factors (Monthly): MKT-RF, SMB, HML, and RF
  • Fama-French 5 Factors (Monthly): Adds RMW and CMA to the 3-factor set
  • Momentum Factor (Monthly): UMD (Up Minus Down)
  • 25 Portfolios Formed on Size and Book-to-Market: 5x5 intersection of size and value quintiles
  • Portfolios Formed on Book-to-Market: Decile portfolios sorted on B/M
  • Portfolios Formed on Earnings-to-Price: Decile sorts on E/P ratio
  • Portfolios Formed on Cash Flow-to-Price: Decile sorts on CF/P ratio
  • Portfolios Formed on Dividend-to-Price: Decile sorts on D/P ratio
  • 12 Industry Portfolios: Monthly returns for broad industry groups
  • International Factors (Developed, Emerging, Europe, Japan, Asia Pacific): Regional 3-factor sets

Data Coverage

  • Time Period: July 1926 to present (updated monthly, typically with a one-month lag)
  • Universe: United States equities (primary) plus international developed and emerging markets
  • Frequency: Monthly returns; portfolios reconstituted annually in June
  • Returns: Value-weighted portfolio returns, expressed as decimal monthly returns

Data Processing

Our backend automatically downloads the latest CSV files from the French Data Library, parses them into a structured database, and performs calculations including annualized returns, rolling statistics, Z-scores, percentile ranks, and cumulative return indices. We do not modify the underlying return data from the source. All derived metrics (such as Sharpe ratios, drawdowns, and regime classifications) are computed from the raw monthly returns using standard financial formulas documented on our Methodology page.

Update Schedule and Data Lag

The Kenneth R. French Data Library publishes on its own schedule, with a lag after each month closes. That lag exists because the underlying portfolio construction depends on accounting data that companies report weeks after their fiscal periods end, and on returns data that must be reconciled before it is considered final. As a result, the most recent month shown on this dashboard will normally trail the current calendar month, and this is expected rather than a sign that the site is stale.

A scheduled job checks the library daily for newly published data and refreshes the stored series when a new month appears. The date labelled "last updated" throughout the site refers to the date of the most recent observation in the data, not the time the page was generated. Historical values can also change: the library occasionally revises previously published figures when source data is restated, and because our percentile and z-score calculations use the full history, a revision slightly shifts every derived statistic for that series.

Why This Source

The French library is used as the primary source because it is the same data underlying the published academic literature on factor investing. Thousands of peer reviewed papers are built on these exact series, which means figures shown here are directly comparable to published research rather than to a vendor's proprietary reconstruction. The construction methodology is documented publicly, the data is free to access, and the series extend back to July 1926 for the three-factor set, which is among the longest continuous return records available in finance.

The trade-off is that these are theoretical portfolio returns computed gross of transaction costs, taxes, short borrow costs, and fees. They describe the underlying phenomenon accurately and they overstate what an investor could have captured. Our article on ETFs versus theoretical indexes works through the difference in detail.

Disclaimer

While we strive for accuracy, this dashboard is a visualization and research tool. Users should independently verify any data against the original source before relying on it for research or investment decisions. We are not affiliated with Kenneth R. French, Dartmouth College, CRSP, or Compustat. Data is provided on an "as is" basis. See our Terms of Service for full disclaimers.