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 \(n=6,8,15,20\) and 50. Also, we apply some examples to illustrate the results. Finally, we compare the common estimators of dependence parameter based on empirical MSE.
Copula functions Celebioǧlu–Cuadras copula Gumbel–Barnett distribution Gumbel’s bivariate distribution Negative quadrant dependence U-statistics
This is a preview of subscription content, log in to check access
The authors would like to thank the associate editor and referees for their careful reading and constructive comments that improved presentation of the paper. The research was supported by a grant from Ferdowsi University of Mashhad (No. MS93321JAB).
Amini M, Jabbari H, Mohtashami Borzadaran GR, Azadbakhsh M (2010) Power comparison of independence test for the Farlie–Gumbel–Morgenstern family. Commun Korean Stat Soc 17(4):493–505Google Scholar
Amini M, Jabbari H, Mohtashami Borzadaran GR (2011) Aspects of dependence in generalized Farlie–Gumbel–Morgenstern distributions. Commun Stat Simul Comput 40:1192–1205CrossRefMATHMathSciNetGoogle Scholar
Asadian N, Amini M, Bozorgnia A (2008) Aspects of dependence in Lomax distribution. Commun Korean Stat Soc 15(2):193–204Google Scholar