Abstract & Overview
The Burr type X distribution has been used in analyzing and fitting lifetime data. It has a close relationship with the Weibull, gamma, exponentiated Weibull and exponentiated exponential distributions and shares various characteristics of these distributions. In this paper, a new bivariate Burr type X distribution is introduced. The construction of this distribution is based on the M mixture representation which combines the mixing and copula ideas. The bivariate Burr type X distribution is constructed based on the Gaussian copula. Bayesian estimation via Gibbs sampler is conducted under different loss functions. A numerical comparison is given for illustrative purpose.