Bayesian and maximum likelihood inference approaches for the discrete generalized Sibuya distribution with censored data

This paper presents inferences under classical (maximum likelihood, ML) and Bayesian approaches for the parameters of the generalized Sibuya (GS) probability distribution considering complete and right censored lifetime data. Under a Bayesian approach, the joint posterior probability distributions of interest are estimated using Markov Chain Monte Carlo (MCMC) simulation methods. A comprehensive simulation study is carried out to assess the performance of the estimation procedure. The usefulness of the GS model is also assessed with applications to two real data sets. Despite its merits, one limitation of the generalized Sibuya distribution is that it does not present great flexibility of fit of the hazard function as compared to other existing lifetime models.