Too Many Factors! Do We Need Them All?

Abstract

We investigate more than a dozen of factors formed on firm characteristics and risk measures that have been claimed to be able to explain cross-sectional asset returns in the literature. In accordance to Fama and French (1993, 1996a), we use these factors in asset pricing, and show that the market portfolio, liquidity and coskewness explain the stock returns as well as the famous Fama-French three factors with momentum. In particular, in most sample periods tested, individual stocks' alphas are insignificant with only two factors, market portfolio and liquidity; in addition, many factors are redundant in asset pricing and are likely to come from data-mining.

Keywords: Asset Price Testing, Average F test, Data-Mining, Empirical Factors

JEL Classification: G12, G14