基于近邻择优策略的股市羊群行为网络协同演化模型及仿真

The Co-Evolution Model of Herding Behavior in Stock Market and its Simulation Based on the Strategy of Neighbor Preferred in Network

基于股市投资者的有限理性特征,从投资主体的自我偏好和邻居主体的行为效应两个维度,构建近邻择优策略下的股市羊群行为网络动态协同演化模型,并通过理论分析和模拟仿真的方法来系统解构股市羊群行为演变的动力学特征。股市网络结构和邻居行为效应影响投资者的行为演化过程;近邻择优函数的凹凸性直接影响到投资者投资行为的动力均衡稳定;投资者网络的连通度和异质性与投资者投资行为均实现协同演变效应;投资者的初始状态基本不影响投资行为的动态均衡。文中建立的协同演化模型及研究结论阐述了影响股市羊群行为演化的特征要素,分析了动态环境下股市羊群行为网络协同演变的规律性特征,为金融管理部门制定宏观政策起到了一定理论支撑。

The herding behavior of stock market system plays an important role in stabilizing the stock market, and has a direct influence in the development of stock market. As a result, the behavior has received increasing interest from the academic and the government. This study considers the inherent characteristics of investors＇ imitation and categorizes the herding behavior evolution of stock market into of the two dimensions -- investors＇ self-psychological preference and the effect of neighbors＇ behavior -- based on behavioral finance theory and complex network theory. Meanwhile, theoretical and simulation analyses are conducted in the following sections in order to provide theoretical support for better understanding the herding behavior of stock market. In the first part, in consideration of investors＇ limited rationality and incomplete features of accessing information in stock markets the coevolution mechanism of herding behaviors is designed under the strategy of neighboring preferred in network. Important factors that can affect an investor＇s trading behavior are classified into two categories： self-psychological preference and neighboring behavior effect. In addition, we systematically analyze the equilibrium state of investors＇ trading behavior by a mean-field equation. In the third part, in order to understand the differences of investors ＇ rationality we conducted a theoretical analysis on the coevolution of trading behaviors＇ equilibrium, affected by the convex and concave of the neighboring effect function. At last, how the network structure among investors, neighboring preferred degree and the initial behavior states of investors affect the evolvement of investors＇ trading behavior is described through the simulation analysis. The results show that the coevolution of investor＇ behavior is affected by the network structure of stock market and the effect of neighboring preference. The stability of equilibrium states of investors＇ behavior dynamics is directly related with the concavity and convexity of the neighbors preference function. Connectivity and heterogeneity of the network play an important role in the coevolution of the investment behavior in stock market. The coevolution equilibrium is barely affected by the initial state of investors＇ behavior. In summary, our proposed model can help understand important factors that would affect investors＇ herding behaviors and provide academic and practical implication for effective financial policies on managing these behaviors.