Π-CAPM: An Asset Pricing Model with Probability Weighting

Abstract

We study the asset pricing implications of probability weighting—the idea that investors overweight rare, high impact events—by generalizing the mean-variance framework to allow for tail overweighting. Our model allows for a unique and homogenous pricing equilibrium with multiple investors and several correlated assets. We find that even a symmetric probability distortion has asymmetric pricing implications. For example, while the price of a left-skewed asset increases in skewness and decreases in variance, the price of a right-skewed asset may not. We also obtain and empirically validate several novel implications on the option-implied premiums on variance, skewness, and correlation.