Industrial Effects and the CAPM: From the Views of Robustness and Longitudinal Data Analysis

The traditional approach by Fama and Macbeth (1973) to the validity of an asset pricing model suffers from two drawbacks. Firstly, it uses ordinary least squares (OLS) method, which is sensitive outliers, estimate time-series beta. Secondly, takes averages slope coefficients cross-sectional regressions ignore importance properties. In this article, robust estimators a longitudinal are applied avoid problems these kinds. We use data on electronics industry in Taiwan’s stock market during period September 1998 December 2001 order examine whether betas Capital Asset Pricing Model (CAPM) valid measure risk industries firms belong explain excess returns. methods we propose lead more explanatory power than OLS results.