Reference analysis for Birnbaum-Saunders distribution

In this paper we consider the Bayesian estimators for the unknown parameters of the Birnbaum-Saunders distribution under the reference prior. The Bayesian estimators cannot be obtained in closed forms. An approximate Bayesian approach is proposed using the idea of Lindley and Gibbs sampling procedure is also used to obtain the Bayesian estimators. These results are compared using Monte Carlo simulations with the maximum likelihood method and another approximate Bayesian approach Laplace's approximation. Two real data sets are analyzed for illustrative purposes.