我国股票市场周期性破灭型泡沫检验——基于门限自回归模型  

Test about Periodically Explosive Bubbles in Chinese Stock Market

作　　者：曾令华[1] 彭益[1] 陈双[1] 

机构地区：[1]湖南大学金融学院,湖南长沙410079

出　　处：《湖南大学学报（社会科学版）》2010年第4期54-57,共4页Journal of Hunan University(Social Sciences)

摘　　要：根据Evans周期性破灭泡沫理论,利用了Enders和Siklos改进后的门限自回归模型,采用了Chan的条件最小二乘法对参数进行了估计,对股票市场泡沫进行了检验。样本区间为1996年1月到2009年2月,结果表明,上证综合指数月度数据的变化可以划分为存在泡沫和不存在泡沫的时期,且两个时期呈现交替出现的状态,我国股票市场存在周期性破灭的泡沫。On the basis of Evan’s periodically collapsing bubbles,this paper uses Ender and Siklos’s improvement of threshold autoregressive model and employs Chan’s conditional least square method in order to make a test to detect the existence of bubbles in stock market. This paper adopts monthly closing price from January 1996 to February 2009. Test results show that Shanghai Composite Index on monthly changes can be divided into two states： the bubble exists and the bubble does not exist,and these two states alternate. This indicates that periodically collapsing speculative bubbles exists in our country’s stock market.

关 键 词：周期性破灭泡沫 门限自回归模型 条件最小二乘法 

分 类 号：F830.91[经济管理—金融学]