BAYESIAN PREDICTION OF PROGRESSIVELY FIRST-FAILURE-CENSORED ORDER STATISTICS BASED ON k-RECORD VALUES FROM WEIBULL DISTRIBUTION

Abstract

Prediction on the basis of censored data has an important role in lifetime studies. This paper discusses Bayesian two-sample prediction of progressively firrst-failure-censored order statistics coming from a future sample based on observed k-record values from two-parameter Weibull distribution. Bayesian interval predictions are obtained based on a continuous-discrete joint prior for the unknown two parameters. Moreover, Bayesian point predictors are investigated under symmetric and asymmetric loss functions. Finally, the estimated risks of various point predictors obtained are compared using the Monte Carlo method.