Non-Bayesian and Bayesian estimation for Lomax distribution under randomly censored with application

Abstract

This paper centers on the examination of the Lomax distribution in the context of randomly censored data. Our primary objectives include deriving maximum likelihood estimators and constructing confidence intervals based on the Fisher information matrix for the unknown parameters in the context of randomly censored data. Furthermore, we develop Bayes estimators utilizing gamma priors, considering both squared error and general entropy loss functions. We also calculate Bayesian credible intervals for the parameters. To offer practical insights, we apply these methods to a real-world dataset subject to random censorship. Finally, for comparative purposes, we conduct a Monte Carlo simulation to assess the various estimation techniques introduced in this study.