Cover Image

Empirical Bayes estimation and its superiority for two-way classification model

Consider a two-way classification model, and let [alpha] and [beta] be the vectors of treatment effects of two factors A and B under investigation. We discuss the problem of constructing Bayes and empirical Bayes (EB) estimators of the linear functions of [alpha] and [beta]. Under general conditions, EB estimators are found to have smaller mean square error matrix than the least sum of squares solutions.
Year of publication:
2003
Authors: Wei, Laisheng ; Chen, Jiahua
Published in:
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 63.2003, 2, p. 165-175
Publisher: Elsevier
Keywords: Bayes estimator Empirical Bayes estimator Mean squared error matrix criterion Two-way classification model
Type of publication: Article
Source:
Persistent link: https://www.econbiz.de/10005223833
Similar items by person