Classical and Bayesian inferential approaches using Lomax model under progressively type-I hybrid censoring

Abstract

In this article, we consider the problem of estimation and prediction on unknown parameters of a Lomax distribution when the lifetime data are observed in the presence of progressively type-I hybrid censoring scheme. In the classical scenario, the Expectation–Maximization (EM) algorithm is utilized to derive the maximum likelihood estimates (MLEs) for the unknown parameters and associated confidence intervals. Under the Bayesian framework, the point estimates of unknown parameters with respect to different symmetric, asymmetric and balanced loss functions are obtained using Tierney–Kadane’s approximation and Markov Chain Monte Carlo (MCMC) technique. Also, the highest posterior density (HPD) credible intervals for the parameters are reckoned using importance sampling procedure. Simulation experiments are performed to compare the different proposed methods. Further, the predictive estimates of censored observations and the corresponding prediction intervals are also provided. One real-life data example is presented to illustrate the derived results.