Bayesian inference of survival curve from record‐breaking observations: Estimation and asymptotic results

Ram C. Tiwari and Jyoti N. Zalkikar

Naval Research Logistics (NRL), 1991, vol. 38, issue 4, 599-609

Abstract: In a variety of industrial situations experimental outcomes are only record‐breaking observations. The data available may be represented as X1, K1., X2, K2,…, where X1, X2,… are the successive minima and K1, K2, … are the number of trials needed to obtain new records. Samaniego and Whitaker [11, 12] discussed the problem of estimating the survival function in both parametric and nonparametric setups when the data consisted of record‐breaking observations. In this article we derive nonparametric Bayes and empirical Bayes estimators of the survival function for such data under a Dirichlet process prior and squared error loss. Furthermore, under the assumptions that the process of observing random records can be replicated, the weak convergence of the Bayes estimator is studied as the number of replications grows large. The calculations involved are illustrated by adopting Proschan's [9] data on successive failure times of air conditioning units on Boeing aircraft, for our purpose. The nonparametric maximum likelihood estimators of the survival function for different choices of the prior are displayed for comparison purposes.

Date: 1991
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://doi.org/10.1002/1520-6750(199108)38:43.0.CO;2-C

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:wly:navres:v:38:y:1991:i:4:p:599-609

Access Statistics for this article

More articles in Naval Research Logistics (NRL) from John Wiley & Sons
Bibliographic data for series maintained by Wiley Content Delivery ().