COMPARISON OF SOME BAYES' ESTIMATORS FOR THE WEIBULL RELIABILITY FUNCTION UNDER DIFFERENT LOSS FUNCTIONS

Huda. Rasheed; Tasnim Al-Baldawi; Nadia Al-Obedy

doi:10.24996/

Authors

Huda. Rasheed Departement of Math., College of Sciences University of Al-mustansiriya, ,.Baghdad-Iraq
Tasnim Al-Baldawi Departement of Math., College of Education, University of Al-mustansiriya, ,. Baghdad-Iraq
Nadia Al-Obedy Departement of Math., College of Sciences University of Al-mustansiriya, ,.Baghdad-Iraq

DOI:

https://doi.org/10.24996/

Keywords:

COMPARISON, ESTIMATORS

Abstract

Weibull distribution is one of the most widely used distributions of
lifetime in reliability engineering. It has the ability to provide reasonably
accurate failure analysis and failure forecasts especially with extremely
small samples. Bayesian approach has received a lot of attention along
with the traditional methods of estimation such as maximum likelihood
and moment estimation methods. The object of the present paper is to
compare some Bayes' estimators for the scale Parameter of the Weibull
reliability function using different loss functions, based on Jeffrey prior
information for estimating the scale parameter of Weibull distribution.
The comparison was based on a Monte Carlo study. Through the
simulation study comparison was made on the performance of these
estimators with respect to the mean square error (MSE) and the mean
percentage error (MPE). The results of comparison by MSE showed
that the Weibull reliability function based upon squared error loss
function was the best followed by the modified El- Sayyad's loss
function. While comparison by MPE showed the reverse.