Bayesian Solutions for the Factor Zoo: We Just Ran Two Quadrillion Models

We propose a novel framework for analyzing linear asset pricing models: simple, robust, and applicable to high-dimensional problems. For (potentially misspecified) stand-alone model, it provides reliable price of risk estimates both tradable nontradable factors, detects those weakly identified. competing factors (possibly nonnested) models, the method automatically selects best specification—if dominant one exists—or Bayesian model averaging–stochastic discount factor (BMA-SDF), if there is no clear winner. analyze 2.25 quadrillion models generated by large set find that BMA-SDF outperforms existing in- out-of-sample.