EXTENDED HOMOGENEOUS PROCESSES AND BAYES ESTIMATION OF RELIABILITY FUNCTIONS
Main Article Content
Abstract
The problem of estimation a reliability function is established in the Bayesian nonparametric context; however parametric techniques are used. Extended homogeneous prncesses are defined whose sample paths may be assumed to be increasing hazard rates by properly choosing the parameter functions of the processes. Estimators are obtained in the mentioned processes and their asymptotic properties are studied. An application for simulated dada is given.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
References
Dykstra, H. L. and Laud, P., "A Bayesian nonparametric approach to reliability", Ann. Statist. 9, n. 2, 356-367, 1981.
Doksum, K., "Tailfree and neutral random probabilities and their posterior distributions", Ann. Probability 2, 183-201, 1974.
Ferguson, T . S., "A Bayesian analysis of some nonparamet.ric prnblems", Ann. Statist. 1, 209-230, 1973.
Morales, D., Quesada, V. and Pardo, L., "Estimadon parametrica bayesiana no parametrica de
funciones de supervivencia con observaciones parcialmentc censuradas", Trabajos de Estadistica 1, n. 1, 70-87, 1986.
Morales, D., Paido, L. and Quesada, V., "Estimation of a survival function with doubly censored data and Dirichlet process prior knowledge on the observable variable", Comm. in Statist. (Sim. and comp.) I, 349-362, 1990.