Confidence intervals for reliability functions of an exponential distribution under random censorship. (English) Zbl 0717.62025

Summary: The asymptotic normality of Bayes estimators of the reliability function of an exponential distribution based on randomly censored data is studied. A Monte-Carlo simulation is used to examine how well two large- sample confidence bands for Bayes estimators do in small and moderate samples. The results are compared with the confidence intervals for the maximum likelihood estimators.


62F12 Asymptotic properties of parametric estimators
62F15 Bayesian inference
62F25 Parametric tolerance and confidence regions
62E20 Asymptotic distribution theory in statistics
65C05 Monte Carlo methods
62N05 Reliability and life testing
Full Text: EuDML Link


[1] J. D. Emerson: Effects of censoring on the robustness of exponential-based confidence intervals for medial lifetime. Commun. Statist. BIO (1981), 6, 617-627.
[2] J. Hurt: On estimation in the exponential case under random censorship. Proc. Third Pannonian Symp. on Math. Statist., Visegrad, Hungary, Akademiai Kiado, Budapest 1982. · Zbl 0535.62031
[3] J. Hurt: Comparison of some reliability estimators in the exponential case under random censorship. Proc. Fifth Pannonian Symposium on Math. Statist., Visegrad, Hungary, Akademiai Kiad6, Budapest 1986. · Zbl 0662.62102
[4] J. A. Koziol, S. B. Green: A Cramer - von Mises statistic for randomly censored data. Biometrika 63 (1976), 465-474. · Zbl 0344.62018
[5] R. G. Miller, Jr.: Survival Analysis. Wiley, New York 1981. · Zbl 0589.62092
[6] C. R. Rao: Linear Statistical Inference and Its Applications. Wiley, New York 1973. · Zbl 0256.62002
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.