On Randomly Censored Generalized Inverted Exponential Distribution

SYNOPTIC ABSTRACT

Life testing experiments are conducted to collect lifetime information on patients in survival analysis. The subjects of testing in reliability theory are electronic, electrical, and mechanical devices. Because of time and cost constraints, it is difficult to collect lifetime data on all items, and therefore, the experiment is terminated before its completion. Various types of censoring schemes are used by the practitioners. Random censoring is an important censoring scheme in which the time of censoring is not fixed but taken as random. In this article, we study the generalized inverted exponential distribution under random censoring. Maximum likelihood estimators of the parameters and expected Fisher information under random censoring model are derived. Bayes estimators of the parameters under squared error loss function are obtained using Lindley's approximation and importance sampling methods. Also, highest posterior density credible intervals of the parameters based on importance sampling procedure are constructed. A Monte Carlo simulation study is conducted to compare the performance of various estimators developed. A randomly censored real dataset illustrates the estimation procedures developed in this study.

KEY WORDS AND PHRASES:

• Bayes estimation
• expected Fisher information
• importance sampling
• Lindley approximation
• maximum likelihood estimation
• random censoring

Acknowledgments

The authors are thankful to the Associate Editor and anonymous referees for their valuable comments and suggestions about the earlier version of the manuscript, which led to substantial improvement in this article.