A Bayesian Approach to the Comparison of Two Means

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

A hierarchical Bayesian approach to the problem of comparison of two means is considered. Hypothesis testing, ranking and selection, and estimation (after selection) are treated. Under the hypothesis that two means are different, it is desired to select the population which has the larger mean. Expressions for the ranking probability of each mean being the larger and the corresponding estimate of each mean are given. For certain priors, it is possible to express the quantities of interest in closed form. A simulation study has been done to compare mean square errors of a hierarchical Bayesian estimator and some of the existing estimators of the selected mean. The case of comparing two means in the presence of block effects has also been considered and an example is presented to illustrate the methodology.

Original languageEnglish (US)
Pages (from-to)3785-3800
Number of pages16
JournalCommunications in Statistics - Theory and Methods
Volume18
Issue number10
DOIs
StatePublished - Jan 1 1989

Fingerprint

Bayesian Approach
Ranking and Selection
Bayesian Estimator
Hypothesis Testing
Mean square error
Ranking
Closed-form
Express
Simulation Study
Estimator
Methodology
Estimate

All Science Journal Classification (ASJC) codes

  • Statistics and Probability

Cite this

@article{cb7d326bae164f1d866b54928384373a,
title = "A Bayesian Approach to the Comparison of Two Means",
abstract = "A hierarchical Bayesian approach to the problem of comparison of two means is considered. Hypothesis testing, ranking and selection, and estimation (after selection) are treated. Under the hypothesis that two means are different, it is desired to select the population which has the larger mean. Expressions for the ranking probability of each mean being the larger and the corresponding estimate of each mean are given. For certain priors, it is possible to express the quantities of interest in closed form. A simulation study has been done to compare mean square errors of a hierarchical Bayesian estimator and some of the existing estimators of the selected mean. The case of comparing two means in the presence of block effects has also been considered and an example is presented to illustrate the methodology.",
author = "Duncan Fong",
year = "1989",
month = "1",
day = "1",
doi = "10.1080/03610928908830123",
language = "English (US)",
volume = "18",
pages = "3785--3800",
journal = "Communications in Statistics - Theory and Methods",
issn = "0361-0926",
publisher = "Taylor and Francis Ltd.",
number = "10",

}

A Bayesian Approach to the Comparison of Two Means. / Fong, Duncan.

In: Communications in Statistics - Theory and Methods, Vol. 18, No. 10, 01.01.1989, p. 3785-3800.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Bayesian Approach to the Comparison of Two Means

AU - Fong, Duncan

PY - 1989/1/1

Y1 - 1989/1/1

N2 - A hierarchical Bayesian approach to the problem of comparison of two means is considered. Hypothesis testing, ranking and selection, and estimation (after selection) are treated. Under the hypothesis that two means are different, it is desired to select the population which has the larger mean. Expressions for the ranking probability of each mean being the larger and the corresponding estimate of each mean are given. For certain priors, it is possible to express the quantities of interest in closed form. A simulation study has been done to compare mean square errors of a hierarchical Bayesian estimator and some of the existing estimators of the selected mean. The case of comparing two means in the presence of block effects has also been considered and an example is presented to illustrate the methodology.

AB - A hierarchical Bayesian approach to the problem of comparison of two means is considered. Hypothesis testing, ranking and selection, and estimation (after selection) are treated. Under the hypothesis that two means are different, it is desired to select the population which has the larger mean. Expressions for the ranking probability of each mean being the larger and the corresponding estimate of each mean are given. For certain priors, it is possible to express the quantities of interest in closed form. A simulation study has been done to compare mean square errors of a hierarchical Bayesian estimator and some of the existing estimators of the selected mean. The case of comparing two means in the presence of block effects has also been considered and an example is presented to illustrate the methodology.

UR - http://www.scopus.com/inward/record.url?scp=0011584602&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0011584602&partnerID=8YFLogxK

U2 - 10.1080/03610928908830123

DO - 10.1080/03610928908830123

M3 - Article

AN - SCOPUS:0011584602

VL - 18

SP - 3785

EP - 3800

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

SN - 0361-0926

IS - 10

ER -