Behavioral Biases as Drivers of Complexity in Stock Markets: An Agent-Based Modeling Approach

By modeling financial systems as Complex Adaptive Systems, this study investigates how behavioral biases influence emergent complexity in stock markets. The study integrates heterogeneous agents, such as rational traders, herding agents, overconfident traders, and anchoring/disposition-driven investors, within a Limit Order Book framework calibrated to both U.S. and Indian market conditions using an Agent-Based Modeling (ABM) approach implemented through the high-fidelity ABIDES simulation environment. Price dynamics, volatility patterns, and liquidity structures were analyzed by Monte Carlo simulation experiments with different behavioral compositions. The results show that behavioral biases cause nonlinear price reactions, produce heavy-tailed return distributions that distort order-book complexity, and greatly increase volatility. The market shifts from a stable, rational regime to a highly volatile, complex regime characterized by contagion and fragile liquidity as the proportion of biased actors rises. Overall, the findings show that the complexity and systemic instability of emerging markets are primarily driven by behavioral heterogeneity.