Bivariate Basu-Dhar geometric model for survival data with a cure fraction

Under a context of survival lifetime analysis, we introduce in this paper Bayesian and maximum likelihood approaches for the bivariate Basu-Dhar geometric model in the presence of covariates and a cure fraction. This distribution is useful to model bivariate discrete lifetime data. In the Bayesian estimation, posterior summaries of interest were obtained using standard Markov Chain Monte Carlo methods in the OpenBUGS software. Maximum likelihood estimates for the parameters of interest were computed using the \textquotedblleft maxLik" package of the R software. Illustrations of the proposed approaches are given for two real data sets.