Research‎ > ‎

Elephants and the Cross-Section of Expected Returns

Standard GMM cross-sectional asset pricing tests are susceptible to a trade-off: They can generate high explanatory power for factor models by allowing the estimated factor means to substantially deviate from the observed sample averages of the factors. In fact, by shifting the weights on the moment conditions, any desired level of cross-sectional fit can be attained. This property is a feature of the GMM estimation design and works for weak as well as strong factors, and for all sample sizes and test assets. To substantiate the trade-off, we run placebo tests based on simulated as well as empirical data.