סמינר במימון חשבונאות
Speaker: Doron Avramov, IDC
Integrating Factor Models
Joint with Si Cheng (Chinese University of Hong Kong) , Lior Metzker (HUJI) and Stefan Voigt (University of Copenhagen and Danish Finance Institute)
This paper develops a comprehensive framework to address uncertainty about the correct factor model specification. Asset pricing inferences draw on a composite model that integrates over competing factor models weighted by posterior probabilities. The analyses show that models with time-invariant parameters record near-zero probabilities, and post-earnings announcement drift, quality-minus-junk, and intermediary capital factors are incremental to the market. From an investment perspective, augmenting the model disagreement about expected returns and the probability weighted uncertainty about model parameters, a Bayesian agent perceives equities as considerably riskier than sample estimates. In addition, the integrated model delivers stable strategies, even during market downturns.