A Comparison of New Factor Models

Abstract

This paper compiles an extensive data library with 437 anomaly variables. Controlling for microcaps leads to 161 significant anomalies with NYSE breakpoints and value-weighted returns and 216 with all-but-micro breakpoints and equal-weighted returns. Liquidity is largely insignificant. The q-factor model has the lowest average magnitude of (and the lowest number of significant) high-minus-low alphas among all the models. The q-factor model outperforms a competing five-factor model in explaining momentum and profitability anomalies. Fundamentals, including investment and profitability, not liquidity, are the key driving forces of the broad cross section of average stock returns.