Do risk exposures explain accounting anomalies? A new testing method

Abstract I quantitatively evaluate how much of the cross‐sectional return predictive ability a range accounting anomalies can be attributed to firm‐specific stock covariances with market risk factors. Using novel two‐step regression‐based testing method, find robust evidence that factor exposures do not explain meaningful fraction time‐series variations in coefficients anomaly variables. Out‐of‐sample tests further confirm it is mispricing component mainly responsible for predictability.