Integrating Factor Models

This paper develops a comprehensive framework to address uncertainty about the correct factor model. Asset pricing inferences draw on composite model that integrates over competing models weighted by posterior probabilities. Evidence shows unconditional record near-zero probabilities, while post-earnings announcement drift, quality-minus-junk, and intermediary capital are potent factors in conditional asset pricing. Out-of-sample, integrated performs well, tilting away from subsequently underperforming factors. Model makes equities appear considerably riskier, disagreement expected returns spikes during crash episodes. Disagreement spans all return components involving mispricing, loadings, risk premia. article is protected copyright. All rights reserved