Macroeconomic Content of Characteristics-Based Asset Pricing Models: A Machine Learning Analysis

We consider five characteristics-based asset pricing models and study whether the non-market components of their stochastic discount factors (SDFs) are associated with macroeconomic shocks. Our analysis involves a comprehensive set of 127 macroeconomic variables and uses machine learning techniques to mitigate the overfitting problem caused by a large number of explanatory variables. We find that macroeconomic shocks are totally unrelated to the non-market components of the SDFs. This conclusion extends to several theory-motivated macroeconomic factors. Thus, our results suggest that the empirical success of characteristics-based asset pricing models is produced by their ability to identify behavioral factors in stock returns.