Chi-Squared Type Test for the AFT-Generalized Inverse Weibull Distribution

Abstract

The generalized inverse Weibull distribution is a newlife time probability distribution which can be used to model a variety of failure characteristics. It has several desirable properties and nice physical interpretations which enable them to be used frequently. In this article, we present a chi-squared goodness-of-fit test for an accelerated failure time (AFT) model with generalized inverse Weibull distribution (GIW) as the baseline distribution, in both of complete and censored data. This test is based on a modification of the NRR (Nikulin-Rao-Robson) statistic Y², proposed by Bagdonavicius and Nikulin (2011), for censored data. Two applications of real data are given to illustrate the potentiality of the proposed test.

Keywords:

• Accelerated failure time model
• Censored data
• Maximum likelihood estimation
• NRR statistic.

Mathematics Subject Classification:

• 60E15
• 62E10
• 62F12
• 62G10
• 62N05.

Acknowledgment

The authors are grateful to Professor M. Nikulin and the referees for their constructive comments and suggestions which helped improve the presentation of this article greatly.