The Dynamic Interplay of Behavioral Biases and Algorithmic Trading: An Agent-Based Modeling Approach to Market Efficiency

Abstract

This paper investigates the complex relationship between behavioral biases exhibited by human traders and the increasing prevalence of algorithmic trading systems in financial markets. Utilizing an agent-based modeling (ABM) framework, we simulate a market environment populated by both behavioral and algorithmic agents. Behavioral agents are endowed with cognitive biases, such as loss aversion, herding, and anchoring, while algorithmic agents are programmed with rational strategies and high-frequency trading capabilities. The simulation results demonstrate that the interaction between these two agent types significantly impacts market efficiency, volatility, and price discovery. Specifically, we find that algorithmic trading can exacerbate the effects of behavioral biases, leading to increased market instability, but also possesses the potential to mitigate these biases under certain market conditions. This research contributes to a deeper understanding of the evolving dynamics of financial markets in the age of algorithmic dominance and provides insights for policymakers and market participants seeking to enhance market stability and efficiency.