Pricing anomalies in a general equilibrium model with biased learning

Abstract

We investigate the emergence of momentum and reversal anomalies in a general equilibrium model with complete markets and cognitively biased agents. General equilibrium and market completeness avoid spurious effects due to portfolio composition or price stickiness. Taking inspiration from and merging different strands of empirical literature, we try to identify anomalies in the most general way, studying return autocorrelation patterns, price gaps following sequences of specific events, and relative performances of suitably defined portfolios. We show that these three characterizations are not equivalent. They capture different aspects of mispricing and relate differently to the behavioral characteristics of the agents. Often, similar anomalous patterns struggle to coexist under seemingly related biases. Overall, the model is generically able to reproduce the empirical evidence of momentum profits that subsequently revert.