Comparison between Bayesian and maximum likelihood methods for parameters and the reliability function of Perks distribution

In this paper, we have derived Bayesian estimation for the parameters and reliability function of Perks distribution based on two different loss functions, Lindley’s approximation has been used to obtain those values. It is assumed that the parameter behaves as a random variable have a Gumbell Type...

Full description

Saved in:
Bibliographic Details
Published in:Iraqi journal of science 2018-03, Vol.59 (1B), p.369-376
Main Authors: al-Sultani, Shuruq Ahmad Karim, Muhammad, Sahar Ahmad
Format: Article
Language:English
Subjects:
التوزيع
الدوال
الرياضيات
تحليل بايزي
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we have derived Bayesian estimation for the parameters and reliability function of Perks distribution based on two different loss functions, Lindley’s approximation has been used to obtain those values. It is assumed that the parameter behaves as a random variable have a Gumbell Type  prior with noninformative is used. And after the derivation of mathematical formulas of those estimations, the simulation method was used for comparison depending on mean square error (MSE) values and integrated mean absolute percentage error (IMAPE) values respectively. Among of conclusion that have been reached, it is observed that, the LE-NR estimate introduced the best perform for estimating the parameter λ.
ISSN:0067-2904
2312-1637
DOI:10.24996/IJS.2018.59.1B.15