A Partially Rational Model for Financial Markets: The Role of Social Interactions on Herding and Market Inefficiency

Abstract

This work investigates how social influence affects the collective behavior of interconnected financial agents in an artificial market. Each agent bases her trading decisions on her perceived value of the traded asset. If the interconnections between agents are not considered, an efficient market emerges, where the intrinsic value of the traded asset is correctly estimated. In the presence of social interactions, modeled through a scale-free network, the trading decisions of each agent also depends on the perception her neighbors have on the asset value. We illustrates how sociality can yield herding, which in turn degrades market efficiency and stability. Then, we propose a control strategy to mitigate herding so as to reduce volatility and regain market efficiency.