基于pair copula的多维股市尾部相关性分析

Analysis on Tail Dependence of Multivariate Stock Market Based on Pair copula

采用pair copula模型对多维情形下的尾部相关性进行了分析,以中国内地及周边国家和地区的股市周收盘价为研究对象,采用经验分布拟合边缘分布,引入藤结构,并结合t-copula、Clayton copula和Joe-Clayton copula分解多维密度函数,结果表明,pair copula模型可以很好地解决多维情况下的尾部相关性分析。

Adopting pair copula model and taking the weekly--closing--price of the mainland of China and neighboring countries and regions as the research object, this paper analyzes the tail dependence in multi--di- mensional case. Empirical distribution function is applied to fit marginal distribution, and vine structure is introduced to decompose the multivariate density function with t--copula, Clayton copula and Joe--Clayton copula. The results show that pair copula methodology can solve the tail dependence in multi--dimensional case efficiently.