基于财富与信息角度的人工股票市场建模及非线性特征形成机理  

Modeling the Artificial Stock Market Based on Wealth and Information and Study the Formation Mechanism of Nonlinear Characteristic

作　　者：邹琳[1] 马超群[1] 刘钰[1] 崔璨[1] 

机构地区：[1]湖南大学工商管理学院,湖南长沙410082

出　　处：《系统工程》2010年第10期29-35,共7页Systems Engineering

基　　金：国家杰出青年科学项目(70825006);教育部高校博士点基金资助项目(200805320024);湖南省社会科学基金资助项目(09YBB085)

摘　　要：在Chinrella人工股票市场的交易框架下,建立了在异质信念模块中增加财富与信息模型的人工股票市场。并通过多次实验,研究了财富与信息对股票市场非线性特征的影响。结果表明,收益率的尖峰、厚尾,波动聚集性,长期记忆性这三种非线性特征的形成各不相同。财富分布均匀更多导致收益率的尖峰、厚尾,长期记忆性产生,而财富分布不均匀导致收益率波动聚集性的产生;技术分析者信息处理方面,不均匀处理更多导致尖峰、厚尾产生,均匀处理更多导致波动聚集性产生,利用更多过去信息将更多导致长期记忆性产生。最后,从股票市场非线性特征的形成机理角度提出了完善股票市场的建议。Under trading framework of Chinrella＇s artificial stock market,we build a new artificial stock market to add wealth and information models in heterogeneous belief module.Through many experiments,we study influence of wealth and the information on the nonlinear characteristic of stock market.The results show that the formations of three nonlinear characteristics such as？leptokurtosis,volatility clustering,long memory of returns are different.The even distribution of wealth will more likely to lead to leptokurtosis and long memory of returns,while the uneven distribution of wealth will more likely lead to volatility clustering of returns.When technical analysts process information unequally,leptokurtosis will be more likely to occur,whereas volatility clustering will be more likely to occur.The use of more past information tends to lead to long memory of returns.Finally,we give some suggestion on how to improve stock markets from the view of formation mechanism of the stock market＇s nonlinearity.

关 键 词：人工股票市场 非线性特征 信息 财富 

分 类 号：F830[经济管理—金融学]