Stylized facts in the cross-section of asset returns: a multi-agent system approach

We integrate multi-agent modeling with asset pricing research to investigate stylized facts in the cross-section of returns. Building on the framework of [Schmitt, N., and F. Westerhoff. 2021. “Trend Followers, Contrarians and Fundamentalists: Explaining the Dynamics of Financial Markets.” Journal of Economic Behavior & Organization 192:117–136. https://doi.org/10.1016/j.jebo.2021.10.006.], we introduce a multi-asset dimension where heterogeneous agents select from trend-following, contrarian, and fundamental trading strategies. Our model simulates a financial market with numerous assets and effectively captures simulated counterparts of three major cross-sectional asset-pricing patterns – short-term reversal, momentum, and long-horizon reversal – linking them through a unified mechanism stemming from interactions among strategies. Additionally, it reproduces various stylized facts in the time series, such as bubbles, crashes, excess volatility, fat-tailed return distributions, and volatility clustering.