Abstract
Abstract
In this study, a new technique has been applied for eliciting the prior density function for the empirical Bayes by utilizing the characteristic function, which frees the Bayesian inference from subjectivity. The empirical Bayes estimates have been studied for the Weibull model parameters based on the characteristic prior and the informative gamma prior in terms of the mean squared errors and the mean percentage errors, via Monte Carlo simulations. The simulation results indicated that the empirical Bayes results based on the characteristic prior outperformed the ones based on the informative gamma prior. Finally, a numerical example is given to demonstrate the efficiency of the proposed priors.
Publisher
Research Square Platform LLC
Reference22 articles.
1. Reliability estimation based on general progressive censored data from the Weibull model: Comparison between Bayesian and Classical approaches;Abd-Elrahman AM;METRON-International Journal of Statistics. LXV,2007
2. Bayesian approach to life testing and reliability estimation using asymmetric loss function;Basu AP;J. Statist. Plann. Infer.,1991
3. An engineering approach to Bayes estimation for the Weibull distribution;Calabria R;Microelectron, Reliability,1994
4. Empirical Bayes methods applied to estimating fire alarm probabilities;Carter G;Journal of the American Statistical Association,1974
5. Estimates of income for small places: An application of James-Stein procedures to census data;Fay RE;Journal of the American Statistical Association,1979