基于CopulaACD模型的股票连涨和连跌收益率风险分析.pdf
文本预览下载声明
第30卷第2期 系统工程理论与实践 Vbl.30.No.2
2010年2月 SystemsEngineering—Theory&PracticeFeb.,2010
文章编号:1000-6788(2010)02—0298-07中图分类号:F830.9 文献标志码:A
胡心瀚,叶五一,缪柏其
(中国科学技术大学统计与金融系,合肥230026)
摘要分别使用包含“天数变量”的Log-ACD和Copula模型对股票的连涨和连跌收益率的边
缘分布以及二者的联合分布进行了拟合,检验结果表明该模型拟合的效果要优于传统方法.对上证
180指数数据做了实证研究,并使用条件VaR对股票连涨连跌收益率进行风险分析,实证结果证明
该模型的拟合结果与股市的实际情况是相吻合的.投资者可以依照模型得出的“涨跌风险对比图”
分析当前股票市场的涨跌风险对比,从而指导投资行为.
关键词上证180指数;Copula-ACD模型;条件VaR
Risk of and stock based
analysiscontinuouslyrisingfalling yield
on method
Copula-ACD
HU Wu-yi,MIAO
Xin-han,YE Bai-qi
ofStatisticsand ofScienceand of
Finance,UniversityTechnologyChina,Hefei
(Department 230026,China)
modelwhich the“numberof and modelare
AbstractThe includes
Log-ACD daysvariableCopula
used tofitthe and distributionof and stock
respectivelymarginaljoint continuouslyrisingfalling yield.
Theresultoftestshowsthatthismodelis betterthantraditionalmethods.Inthis
comparatively paper.
isusedfor andconditional眦isrisk
stock180index for
Shanghai empiricalanalysis computedanalysis.
The result thatthismodeliSconsistenttoactualmarket.andinvestorcall of
proves use‘figure
empirical
of
and risk’to andestimatetheiraction
risingfalling analyze investing.
显示全部