Beschreibung:
We examine how the return predictability of deep learning models varies with stocks’ vulnerability to investors’ behavioral biases. Using an extensive list of anomaly variables, we find that the long-short strategy based on deep learning signals generates greater returns for stocks more vulnerable to behavioral biases, i.e., small, young, unprofitable, volatile, non-dividend-paying, close-to-default, and lottery-like stocks. This performance of deep learning models becomes pronounced for stocks held by less sophisticated investors, when investor sentiment is high, and when disagreement is serious. These results suggest that deep learning models with nonlinear structures are useful for capturing mispricing induced by behavioral biases