Empirical Bayes Inference on the Inverse Weibull Model Parameters based on Characteristic Prior

Abstract

Abstract Empirical Bayes has become increasingly popular and has been applied to many types of problems. To increase its popularity, a new method has been used for constructing prior density functions by utilizing the characteristic function. For comparing the empirical Bayes estimates based on the characteristic prior and the informative prior, the mean squared errors and the mean percentage errors for the inverse Weibull distribution parameters have been derived based on symmetric and asymmetric loss functions via Monte Carlo simulations. The simulation results indicated that the empirical Bayes based on the characteristic prior provides a better estimate and outperforms the informative gamma prior for different sample sizes. Moreover, the characteristic prior is flexible for applications without using hyperparameters. Finally, a numerical example is given to demonstrate the efficiency of the proposed priors.