COMPARATIVE STUDY OF U-STATISTIC BASED TESTS FOR EXPONENTIALITY: APPLICATION TO RELIABILITY AND SURVIVAL DATA

Read full text Summarise with VIRAT
Authors: A.G. M. El-Sherbiny, N. A. Abdul Alim, A.K. El-Kholy, Eman S. El-Atfy
Cited by: 0
References: 12

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This study introduces a novel U-test for assessing whether lifetime data conforms to the Exponential Better than Used in Convex Laplace Transform Order (EBUCL) class, a generalization of traditional exponential models in reliability and survival analysis. The proposed test leverages moment inequalities and asymptotic normality to distinguish exponential distributions from their EBUCL counterparts. Through rigorous evaluation using Pitman’s asymptotic efficiency and Monte Carlo simulations, the test demonstrates superior performance, particularly for Weibull and Gamma distributions. Practical applications to real-world datasets, including COVID-19 mortality rates, leukemia survival times, and equipment failure data, highlight the test’s ability to detect subtle deviations from exponentiality. The results underscore the test’s robustness and efficiency, making it a valuable tool for reliability and survival analysis.

Fields of Study & Subjects

Statistical Distribution Estimation and Applications Probability and Statistical Research Reliability and Maintenance Optimization

Publication details

Journal / source
Reliability: Theory & Applications
Publication date
1 January 2026
Language
EN