A family of bayes estimators for doubly censored Weibull distribution. Issue 4 (18th May 2020)

• Record Type: Journal Article
• Authors: Feroze, Navid; Aslam, Muhammad
• Is Part Of: Journal of statistics & management systems. Volume 23:Issue 4(2020)
• Page Start: 737
• Page End: 759
• Publication Date: 2020-05-18
• Subjects: 62F15 and 62N01; Quadrature method -- Gibbs sampler -- importance sampling -- Lindley's approximation -- Tierney and Kadane's approximation -- posterior distributions and loss functions; Statistics -- Periodicals; Mathematical models -- Periodicals; Mathematical models; Statistics; Periodicals; 519.5
• DOI: 10.1080/09720510.2019.1658413
• Languages: English
• ISSNs: 0972-0510

Abstract

Abstract: Among the standard lifetime distributions the Weibull distribution has a supreme place due to its additional advantages over its counterparts. In this paper, we have considered the Bayesian estimation of the Weibull distribution when samples are doubly censored. Two loss functions and two priors have been considered for the estimation of the parameters from the Weibull distribution. However the Bayes estimators cannot be derived in the closed form, therefore we have used five approximation methods namely Quadrature method, Gibbs sampler, importance sampling, Lindley's approximation and Tierney and Kadane's approximation for the approximate estimation of the parameters. The analysis of simulated and real datasets has been considered for the investigation of the performance of the different estimators. Based on the findings from the detailed analysis, it can be concluded that importance sampling and Tierney and Kadane's approximation have better performance with certain conditions.