Universiti Teknologi Malaysia Institutional Repository

Bayesian estimation of reliability

Rahmi, Depriwana (2005) Bayesian estimation of reliability. Masters thesis, Universiti Teknologi Malaysia, Faculty of Science.

[img]
Preview
PDF
436kB

Official URL: http://dms.library.utm.my:8080/vital/access/manage...

Abstract

Most of statistical theory thought at the undergraduate level is based on classical probability or frequentist probability. Also, the estimation techniques widely used are maximum likelihood estimation (MLE) and methods of moment. This dissertation discusses the subjective probability as an alternative theory, where Bayesian estimation is used. Based on a random sample of 20 observations that was generated from Weibull distribution, we compare the reliability estimation using MLE and Bayesian approach. We found out that the mean percentage error for Bayesian is 7.9678, which is lower than MLE mean percentage error of 10.8434. Thus, bayesian reliability estimation is better.

Item Type:Thesis (Masters)
Additional Information:Thesis (Sarjana Sains (Matematik)) - Universiti Teknologi Malaysia, 2005
Uncontrolled Keywords:estimation techniques
Subjects:Unspecified
Divisions:Science
ID Code:34744
Deposited By: Kamariah Mohamed Jong
Deposited On:24 Jul 2017 13:13
Last Modified:11 Oct 2017 06:16

Repository Staff Only: item control page