On Progressively Type-II Censored Inverse Lomax Distribution: Characterizations, Estimation and Application to Cancer Data

Abstract

This paper discusses some additional distributional characteristics of inverse Lomax distribution and proposes the classical and Bayesian estimation procedures of the survival characteristics. The observed sample information is progressively type-II censored with fixed removal. In classical estimation method, maximum likelihood estimators for the parameters are obtained and then using invariance property, maximum likelihood estimators of the considered survival characteristics are computed. The Bayesian aspect has been discussed with gamma prior under generalized asymmetric loss function. Further, different interval estimations such as asymptotic confidence interval, bootstrap confidence interval and the Bayes credible interval have also been constructed. The Monte Carlo simulation has been carried out to compare the point and interval estimates of the considered survival characteristics. The comparisons are made in terms of average mean squared error and corresponding widths of the intervals. To show the practical applicability of the proposed study, two data sets pertaining to cancer patients have been reanalyzed.