Answers are grounded in this paper and cite their sources.
This paper addresses the problem of discriminating among competing lifetime models within the inverse family of distributions, specifically the inverse Weibull and inverted exponentiated variants of the exponential and Rayleigh distributions. Although these distributions can exhibit similar shapes and overlapping probability density functions, they may differ in key characteristics, such as their failure rate functions, motivating effective methods for distinguishing among them. We focus on likelihood‑ratio-based discrimination and several empirical distribution function statistics, namely the Kolmogorov–Smirnov, Kuiper’s V, Cramér–von Mises, and Anderson–Darling statistics. The asymptotic distributions of the log-ratios of maximized likelihoods are derived and used to approximate the probabilities of correct model selection and determine the minimum sample sizes required to achieve a specified selection probability. A comprehensive simulation study assesses the performance of the proposed discrimination procedures under various scenarios. Two real data sets illustrate the practical applicability of the methods. The study is further extended to censored data, providing insights into the impact of censoring on lifetime model discrimination.
No takes yet. Share an insight, caveat, or question.
Diyali et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: