基于策略演化的股票市场元胞自动机模型  

Cellular automata of stock market based on strategy evolution

作　　者：姜爱克[1] 李雪岩[1] 李雪梅[1] 李学伟[1] 

机构地区：[1]北京交通大学经济管理学院,北京100044

出　　处：《北京交通大学学报》2013年第3期154-160,共7页JOURNAL OF BEIJING JIAOTONG UNIVERSITY

基　　金：国家自然科学基金资助项目(7127302)

摘　　要：将基于Moore型邻居的投资者策略演化行为及策略演化效用函数引入投资者预期驱动的股票市场元胞自动机模型,对演化动力系统做了动力学分析,并研究了策略模仿行为对价格的影响.理论分析及计算机仿真结果表明:股票价格呈现基本面价值收敛,验证了动力系统的稳定性;引入基于Moore型邻居的策略演化规则后,投资者的策略选择出现明显的羊群行为,并且导致价格波动异常,价格发现效率降低;收益率峰度与投资者策略模仿强度有关.Evolution of investors＇ strategies based on Moore neighborhood and utility function of strategy evolution are introduced in the cellular automata of stock market which is drived by investors＇ expectation. Dynamic analysis of the system is made, and impact of investors＇ imitating in strategies is analyzed. Theoretical analysis and computer simulation shows as follows. The price converges at the fundamental value, and the stability of the dynamic system is verified. Herd behaviors of investors＇ strategies selecting, unusual volatility of price and poor efficiency of price discovering are appeared after rules of strategies evolution based on Moore neighborhood are introduced. Kurtosis of return is related with the intense of investors＇ imitating in strategies.

关 键 词：股票价格 元胞自动机 策略演化 动力系统 羊群行为 价格发现 

分 类 号：F830.91[经济管理—金融学]