Bayes probability intervals in a load-strength model

In, this paper Bayes procedures for constructing probability intervals on the parameters and on the reliability function, for complete and censored inverse Weibull samples, are given, both when a poor and a more detailed prior information is introduced into the inferential procedure. Via a Monte Carlo simulation the statistical properties of Bayes estimators are compared to those of the ML ones, assuming both correct and uncorrect prior information on the shape parameter. This study has shown that the Bayes procedure outperforms the ML one even when there is only a poor prior information available.