摘要
研究不同市场形态下股票市场的特征性事实对理解股市运行有指导意义。构建了一个具有基础均衡、泡沫均衡、周期和混沌4种市场形态的多主体股市模型,并对其各种形态进行了计算机仿真实验,然后对仿真收益率序列统计分析。研究表明:4种市场形态均存在尖峰肥尾、波动聚集和长期记忆特性,其中以基础均衡状态最为明显,周期和混沌次之,泡沫均衡最弱。
Research of stylized facts of stock market in different market states has guiding meanings on understanding of stock market operation. We construct a multi-agent model of stock market with four kinds of states: fundamental equilibrium, bubble equilibrium, period and chaos in previous work. Computer simulations are performed for four states respectively, statistical analysis on the return series shows: markets of all states present leptokurtosis, volatility clustering and long memory ; fundamental equilibrium market has strongest, followed by period and chaos market, and bubble equilibrium market has weakest stylized facts.
出处
《复杂系统与复杂性科学》
EI
CSCD
2008年第2期62-71,共10页
Complex Systems and Complexity Science
基金
西南大学青年科技基金(SWUQ2006022)
关键词
人工股市
特征性事实
基于主体的计算经济学
artificial stock market
stylized facts
agent-based computational economics