借助于Copula这一工具对在具有联合分布H(x,y)的二维总体(X,Y)经概率积分变换得到的随机变量H(X,Y)的分布函数K(V)给定条件下,求出关于次序统计量X(n)=max{X1,X2,…,Xn}和Y(n)=max{Y1,Y2,…,Yn}经概率积分变换得到的随机变量Hn(X(n),Y(n))的分布函数Kn(v)的方法进行了研究,并找出了Kn(v)不随样本容量n变化的情形,进而得到了较准确地刻画二维R.V相依性的相关性指标Kendall's τ.
[1] NELSEN R B. An introduction to copulas [M]. Lecture Notes in Statistics 1999, 139.
[2] NELSEN. Distribution function of copulas: a class of BIPIT [J]. Statistics & Probability Letters 2001, 54.
[3] GALAMBOS J. The asymptotic theory of extreme order statistics [M]. New York :Krieger Malabar Academic Press 1987.
[4] Roger B, NELSEN. Bounds on bivariate distribution functions with given margins and measures of association [J]. Communications in Statistics-Theory and Methods 2001, 30(06).