In this paper, a multivariate form of generalized exponential distribution (GED) is constructed using the copula introduced by Al-Hussaini and Ateya [1]. The conditional distribution functions and the corresponding conditional density functions are constructed. Also, all parameters are estimated using the maximum likelihood method in case of the trivariate cases. A simulation study is carried out to study the behavior of the mean squared errors. A real data set is introduced and analyzed using the bivariate generalized exponential distribution introduced in this paper and using the bivariate generalized exponential distribution introduced by Kundu and Gupta [2]. A comparison is carried out between the mentioned models based on the corresponding Kolmogorov-Smirnov (K-S) test statistic to emphasize that the bivariate generalized exponential introduced in this paper fits the data better than the other model.

copulas, generalized exponential distribution, multivariate distributions, maximum likelihood estimation, K-S test statistic.