用pair-copula构建高维相依结构,将n维联合密度函数转化为若干个pair-copula密度函数相乘。在pair-copula的选择上,本文构造了能描述非对称尾部相关性的混合copula函数——M-copula。并在实证分析部分用该方法探索了上证市场上四个板块的相依关系,得到了比较理想的结果。 Using pair-copula to construct the high-dimensional dependency structure, the n-dimensional joint density function is transformed into several pair-copula density functions to be multiplied. In the choice of pair-copula, this paper constructs a mixed copula function that can describe the asymmetry tail correlation - M-copula. In the part of empirical analysis, this method is used to explore the dependence of the four plates on the SSE market, and the ideal result is obtained.