http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
A Class of Admissible Estimators of the Mean in a Hierarchical Bayesian Model
Koh, Tae-Wook 전주대학교 산업경영연구소 2000 産經論叢 Vol.18 No.-
Given a p dimensional (p≥4) normal random vector X with unknown mean vector θ and covariance I_(p) where I_(p) is the p×p matrix, the problem is to estimate θ when the loss function is sum of squared error looses. In a hierarchical Bayesian model, we first find three stage Bayes estimators. Then, we derive classes of admissible estimators out of Bayes estimators. Finally, we provide a class of proper Bayes, minimax, and admissible estimators when the p is greater then or equal to 6.
Some Alternative Approaches to Multistage Bayes Estimators
Koh, Tae Wook Jeonju University Press 1999 Jeonju University Journal Vol.6 No.-
This paper presents three stage Bayes estimators of the mean of the multivanate normal distribution with covariance matrix the identity and sum of squared errors loss We first obtain a class of Bayes estimators with respect to three stage priors using the confluent hypergeometric function and then find the modified maximum likelihood estimators in a hierarchical Bayesian model
IMPROVING ON INADMISSBLE ESTIMATORS IN CONTINUOUS EXPONENTIAL FAMILES.
Koh, Tae Wook 전주대학교 산업경영연구소 1990 産經論叢 Vol.- No.7
We consider a formulation of and solution to the problem of improving upon an inadmissible estimator δ° for arbitrary continuous exponential families and squared error loss.
The Blyth Technique of Proving Admissibility.
Koh, Tae Wook,Kim, Yon Hyong 전주대학교 산업경영연구소 2001 産經論叢 Vol.20 No.-
Being a sufficient condition for admissibility, Farrell(1968) obviously provides a tool for verifying the admissibility of a decision rule. The use of this sufficient condition actually predates Stein(1955), first appearing in Blyth(1951). In this paper we discuss the Blyth technique of proving admissibility.
Hierarchical Bayes Estimators of a Multivariate Normal Mean
Koh, Tae-Wook 전주대학교 산업경영연구소 1992 産經論叢 Vol.- No.8
In four or more dismensions we consider the Bayes estimator (the posterior θ?? based on the joint posterior distribution of (??,t,μ)) of ?? with respect to the density f(??,t,μ). The Bayes estimator ??(??) of ?? minimizes ∫^(∞)_(-∞)∫^(1)_(0)∫_(IR^(p))|??-??(x)|^(2)f(??,t,u|??)dθdtdu with respect to ??(??) for each ??.
A note on inadmissibility of the usual estimator for the mean of a multinormal distribution.
Koh, Tae Wook 전주대학교 산업경영연구소 2001 産經論叢 Vol.20 No.-
Charles Stein showed that it is possible to make a uniform improvement on the maximum likelihood estimator in terms of total squared error risk when estimating several parameters from independent normal observations. Our purpose in this article is to review that the inadmissibility of the usual estimator for the mean of a multinormal distribution by Stein.
Koh, Tae-Wook,Choi, Jung-Min,Lee, Sunghun,Yoo, Seunghyup WILEY-VCH Verlag 2010 ADVANCED MATERIALS Vol.22 No.16
<B>Graphic Abstract</B> <P>A highly conductive polymer layer coated on a microstructured indium tin oxide (ITO) electrode is proposed as a simple way to enhance outcoupling in organic light-emitting diodes. The relatively low refractive index of the conductive polymer provides an index contrast between the organic and ITO layers so that structuring of ITO electrodes can result in a significant optical effect. <img src='wiley_img_2010/09359648-2010-22-16-ADMA200903375-content.gif' alt='wiley_img_2010/09359648-2010-22-16-ADMA200903375-content'> </P>