基于BEMD和Copula的沪港股市相关性研究

A study of the interdependence of the Shanghai and Hong Kong stock markets based on Copula functions and BEMD

为深入探讨沪港两股市间相关结构的微观特征,本文在Copula理论的基础上引入二元经验模态分解(BEMD)算法,分别刻画了上证综指和恒生指数日收益率序列之间的整体相关性和微观相关性。研究结果表明,在港股市间整体相关性方面以及经BEMD分解后不同尺度上微观相关性刻画方面,时变SJC Copula函数都能较好地描述两收益率序列之间的相关结构,即沪港股市间存在时变的非对称尾部相关关系。此结果也进一步证实新方法在刻画相关性方面的有效性。

In order to investigate the microscopic features of the interdependence structure of the Shanghai and Hong Kong stock markets,a bivariate empirical mode decomposition（EMD） algorithm,based on the Copula function,is introduced to describe the total and microscopic dependence between the daily return series of the Shanghai Composite Index and the Hang Seng Index. The studies shown that in terms of the total dependence and microscopic dependence after decomposition by the bivariate EMD,the time-dependent SJC Copula can best describe the interdependence structure between two return series,which indicates that the interdependence of the Shanghai and Hong Kong stock markets is time-dependent,and shows an asymmetric tail. In addition,by comparing the results of the total and microcorrelation analyses,this study also proves the validity of the new model for describing the interdependence of the two stock markets.