Your Cart

Your cart is empty

Home  >  Volume 46 (May, 2018)

18. BAYESIAN ESTIMATION OF THE SCALE PARAMETER OF NEW WEIGHTED EXPONENTIAL DISTRIBUTION UNDER DIFFERENT LOSS FUNCTIONS by A.Isah and A .M.Mustapha Volume 46 (May, 2018 Issue), pp139 –144
Sale price: $5.00
Quantity:

BAYESIAN ESTIMATION OF THE SCALE PARAMETER OF NEW WEIGHTED EXPONENTIAL DISTRIBUTION UNDER DIFFERENT LOSS FUNCTIONS

A.Isah and A .M.Mustapha

Department of Statistics, Ahmadu Bello University, Zaria-Nigeria

Abstract

In this study, the Bayesian estimate of the scale parameter of the New Weighted exponential distribution (proposed in earlier research) was derived under the assumption of non-informative priors. The estimators were obtained using squared error loss function and entropy loss function. Extensive Monte Carlo simulation was carried out in order to compare the performance of the estimators obtained under the Bayesian approach with the maximum likelihood estimator based on their mean square error. The maximum likelihood estimator was found to be better than the Bayes estimator obtained under squared error loss function using uniform prior.  It was observed that the best estimate is obtained under entropy loss function using Extended Jeffrey’s prior when the constant is equal to one (r=1.0) where r is a positive real number. 

click here to download full Abstract
Close
Loading...