Expected Bayesian estimation based on generalized progressive hybrid censored data for Burr-XII distribution with applications

Abstract

In the Bayesian estimation method for the parameters of random distributions, the process of selecting hyperparameters for the prior distributions is one of the important and complex matters that determine the efficiency of the estimation. Therefore, researchers have recently been interested in the expected Bayesian (E-Bayes) estimation as a solution to hyperparameter problems. In this paper, we discuss the Bayes and E-Bayes estimation process based on generalized type-I hybrid censored data from Burr-XII distribution. We used symmetric and asymmetric loss functions, such as squared error, Degroot, quadratic, and linear exponential loss functions. All of these methods were compared using Monte Carlo simulations, using which mean square errors and average of estimators were calculated. Moreover, real data were used as an applied and illustrative example. Finally, some conclusions were drawn in the concluding comments of this paper.