Extremal dependence of copulas: A tail density approach
Extremal dependence of copulas: A tail density approach
复制标题
联结的极值依赖性：尾部密度方法
DOI：
10.1016/j.jmva.2012.07.005
复制
发表时间：
2013
影响因子：
1.6
通讯作者：
Peiling Wu
中科院分区：
文献类型：
--
作者：
Haijun Li;Peiling Wu
随机向量的极值相关性描述了随机向量的联合概率相对于其边缘的联合概率的尾部行为，通常使用其联结的尾部相关函数进行研究。本文引入了尾密度方法来分析仅由密度指定的联结的极值依赖性。建立了尾密度与规则变化密度之间的关系，明确推导了阿基米德尾密度和t尾密度。尾密度方法对于藤连的极值依赖性分析特别有效，因为藤连的尾密度可以递归地写成二元基线连的尾密度和二元连接连的尾密度的乘积形式。
The extremal dependence of a random vector describes the tail behaviors of joint probabilities of the random vector with respect to that of its margins, and has been often studied by using the tail dependence function of its copula. A tail density approach is introduced in this paper to analyze extremal dependence of the copulas that are specified only by densities. The relation between the copula tail densities and regularly varying densities are established, and the tail densities of Archimedean and t copulas are derived explicitly. The tail density approach becomes especially effective for extremal dependence analysis on a vine copula, for which the tail density can be written recursively in the product form of tail densities of bivariate baseline copulas and densities of bivariate linking copulas.