Multifactor, multi-indicator approach to asset pricing: Method and empirical evidence

Abstract

This paper uses a multifactor, multi-indicator approach to test the capital asset pricing model (CAPM) and the arbitrage pricing theory (APT). This approach is able to solve the measuring problem in the market portfolio in testing CAPM;and it is also able to directly test APT by linking the common factors to the macroeconomic indicators. Our results from testing CAPM support Stambough’s (Journal of Financial Economics, 10, 237–268, 1982) argument that the inference about the tests of CAPM isinsensitive to alternative market indexes. We propose a MIMIC approach to test CAPM and APT. The beta estimated from the MIMIC model by allowing measurement error on the market portfolio does not significantly improve the OLS beta, while the MLE estimator does a better job than the OLS and GLS estimators in the cross-sectional regressions because the MLE estimator takes care of the measurement error in beta. Therefore, the measurement error problem on beta is more serious than that on the market portfolio.