Testing the CAPM with Time-Varying Risks and Returns




    Search for more papers by this author
    • School of Business, The University of Michigan and Department of Economics, The Ohio State University, respectively. For useful comments on earlier versions of this paper, we thank John Abowd, Steve Cecchetti, Joel Hasbrouck, Dick Jefferis, Stan Kon, Roger Kormendi, Paul Richardson, Jay Shanken, René Stulz, and seminar participants at Cornell, the Federal Reserve Banks of Atlanta and Cleveland, Ohio State, and Michigan. The comments of an anonymous referee also led to improvements in the paper. Bodurtha acknowledges financial support from a University of Michigan School of Business summer grant and Citicorp, N. A. All errors are our own.


This paper draws on Engle's autoregressive conditionally heteroskedastic modeling strategy to formulate a conditional CAPM with time-varying risk and expected returns. The model is estimated by generalized method of moments. A CAPM that allows mean excess returns to shift in January survives generalized method of moments specification tests for a number of omitted variables. However, a residual dividend yield component is found to remain in the excess returns of smaller firms. We find significant monthly and quarterly components in the risk premia and beta estimates.