Dependence structure and test of independence for some well-known bivariate distributions

被引:0
|
作者
M. Zargar
H. Jabbari
M. Amini
机构
[1] Ferdowsi University of Mashhad,Department of Statistics, Ordered and Spatial Data Center of Excellence
来源
Computational Statistics | 2017年 / 32卷
关键词
Copula functions; Celebioǧlu–Cuadras copula; Gumbel–Barnett distribution; Gumbel’s bivariate distribution; Negative quadrant dependence; U-statistics;
D O I
暂无
中图分类号
学科分类号
摘要
In this paper, we study the dependence structure of some bivariate distribution functions based on dependence measures of Kochar and Gupta (Biometrika 74(3):664–666, 1987) and Shetty and Pandit (Stat Methods Appl 12:5–17, 2003) and then compare these measures with Spearman’s rho and Kendall’s tau. Moreover, the empirical power of the class of distribution-free tests introduced by Kochar and Gupta (1987) and Shetty and Pandit (2003) is computed based on exact and asymptotic distribution of U-statistics. Our results are obtained from simulation work in some continuous bivariate distributions for the sample of sizes n=6,8,15,20\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n=6,8,15,20$$\end{document} and 50. Also, we apply some examples to illustrate the results. Finally, we compare the common estimators of dependence parameter based on empirical MSE.
引用
收藏
页码:1423 / 1451
页数:28
相关论文
共 50 条