A new goodness of fit test for gamma distribution with censored observations

TL;DR

This paper introduces a novel goodness of fit test for gamma distribution that effectively handles censored data, utilizing fixed point characterization and U-Statistic theory, with validation through simulations and real data examples.

Contribution

It presents a new gamma distribution goodness of fit test that incorporates censored observations using fixed point characterization and U-Statistic theory, with detailed asymptotic analysis.

Findings

Test performs well in simulations

Effective with censored and uncensored data

Validated on real datasets

Abstract

In the present paper, we develop a new goodness fit test for gamma distribution using the fixed point characterization. U-Statistic theory is employed to derive the test statistic. We discuss how the right censored observations are incorporated in the test developed here. The asymptotic properties of the test statistic in both censored and uncensored cases are studied in detail. Extensive Monte Carlo simulation studies are carried out to validate the performance of the proposed tests. We also illustrate the test procedure using several real data sets.