基于copula模型的沪深股市日收益率的相关性研究  

The Research on Sequence of Daily Return Correlation Between Shanghai-Shenzhen Stock Index Based on Copula Function

作　　者：刘喜波[1] 王增[1] 谷艳华[2] 

机构地区：[1]北方工业大学理学院,北京100041 [2]中原工学院信息商务学院,河南郑州451191

出　　处：《数学的实践与认识》2015年第11期101-108,共8页Mathematics in Practice and Theory

基　　金：北京市教育委员会科技发展计划项目(km200810009004)

摘　　要：由于沪深股市收益率具有非线性的特征,本文利用Copula函数从定量的角度刻画了上证综指和深证成指的日收益率序列的相关关系,研究表明,沪深股市日收益率序列呈现出很高的相关性,当沪深两市出现大幅震荡时,两市收益率的协同作用将大幅增强,Gaussian Copula函数更好的刻画了沪深股市收益率之间的秩相关性,Gumbel Copula函数在更好的刻画了两收益率序列的上尾相关性,而Clayton Copula函数在分析两序列的下尾相关性时较为出色,在平方欧氏距离标准下,t-Copula较好的拟合了沪深股市的日收益率序列。This article uses Copula function to measure the correlation of the sequence of daily return rate between The Shanghai Composite Index and the Shenzhen Component Index based on the nonlinearity of daily return rate. The result verifies the high correlation between The Shanghai Composite Index and the Shenzhen Component Index. When the two stock markets show large degree of fluctuation, their series of daily return rate tend to change simultaneously. Gaussian Copula function is appropriate to measures the rank correlation and Gumbel Copula function does better in measuring the upper tail correlation, while Clayton Copula performs well in measuring the lower tail correlation. Under the norm of squared Euclidean distance, t-copular fits well in measuring the correlation of the two series of daily return rate.

关 键 词：COPULA函数 秩相关 尾部相关 

分 类 号：F224[经济管理—国民经济] F832.51