|
Activity Number:
|
524
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #305200 |
|
Title:
|
Bayesian Estimation of Burr Type XII Distribution from Progressively Type II--Censored Samples
|
|
Author(s):
|
Seongho Song and Younshik Chung*+ and Chansoo Kim and Junghoon Jang
|
|
Companies:
|
University of Cincinnati and Pusan National University and Kongju National University and Korea Food and Drug Association
|
|
Address:
|
30 Jangjun Dong Guemjung Gu, Busan, 609 735, Korea
|
|
Keywords:
|
Burr Type-XII Distribution ; Progressively Type-II Censored Samples ; Fisher Information ; EM-algorithms ; MCMC
|
|
Abstract:
|
In this research, the estimations of parameters based on a progressively Type-II censored sample from a Burr Type-XII distribution are investigated by using maximum likelihood (ML) approach with EM-algorithms as well as Bayesian approach using MCMC. Additionally, some properties of the model are discussed. The exact expression of the expected Fisher information matrix of parameters in the distribution has been obtained. Based on this result, we construct the confidence interval for the parameters using the asymptotic normality of the ML estimator. We compare it to Bayesian credible interval. We also consider the problem of predicting an independent future observation from the same distribution. The Gibbs and Metropolis samplers are used to predict the behavior of future observations from the same distribution.
|
