Bayesian computational methods for estimation of two-parameters Weibull distribution in presence of right-censored data

In this paper we study the performance of the Metropolis-Hastings and Slice sampling algorithms for estimating the Weibull distribution parameters. The Metropolis-Hastings algorithm is developed considering its two main version namely independent Metropolis-Hastings and random walk Metropolis. A numerical simulation study is carried out to understand performance of the three methods and compare their performances with maximum likelihood estimation. The comparison among methods is made in terms of sample root mean square errors and bias. We find that the Random walk Metropolis and Slice sampling present a complementary behaviour and outperforms the Independent Metropolis-Hastings and Maximum Likelihood estimation, specially, for datasets with censored times. The methods are also illustrated on three real datasets