FIRM SIZE, BOOK-TO-MARKET RATIO, AND SECURITY RETURNS - A HOLDOUT SAMPLE OF FINANCIAL FIRMS

Fama and French (1992) document a significant relation between firm si ze, book-to-market ratios, and security returns for nonfinancial firms . Because of their initial interest in leverage as an explanatory vari able for security returns, Fama and French exclude from their analysis financial firms, thus creating a natural holdout sample on which to t est the robustness of their results. We document that the relation bet ween firm size, book-to-market ratios, and security returns is similar for financial and nonfinancial firms. In addition, we present evidenc e that survivorship bias does not significantly affect the estimated s ize or book-to-market premiums in returns. Our results indicate data-s nooping and selection biases do not explain the size and book-to-marke t patterns in returns.