基于Copula函数的沪深股市相关性分析

摘要

The GPD distribution is chosen to describe log return rate of Shanghai and Shenzhen stock market.The correlation between the Shanghai and Shenzhen stock market is described using the binary Copula function , combined with the statistical characteristics of the sample data .The parameters of the Copula function are estimated .The results show that binary t-Copula function can better capture the tail correlation than binary normal Copula function .There is a strong correlation between Shanghai and Shenzhen stock markets , with the linear correlation coefficientbeing 0 .9356 , and the Kendall correlation coefficient being 0 .7611 , and the Spearman correlation coefficient being 0.9150.%选用GPD分布分别对沪深股市对数收益率尾部进行描述,结合样本数据的统计特征,选择合适的二元Copula函数对沪深股市对数收益率的相关性进行描述,对二元Copula函数的参数进行估计.结果表明:二元t-Copula函数比二元正态Copula函数更能捕捉沪深股市的尾部相关性.沪深股市存在着较强的相关性,线性相关系数为0.9356,Kendall相关系数为0.7611,Spearman相关系数为0.9150.