CHARACTERIZING SUMS OF SQUARES BY THEIR DISTRIBUTIONS

Authors
SEELY JF BIRKES D LEE Y
Citation
Jf. Seely et al., CHARACTERIZING SUMS OF SQUARES BY THEIR DISTRIBUTIONS, The American statistician, 51(1), 1997, pp. 55-58
Citations number
11
Categorie Soggetti
Statistic & Probability","Statistic & Probability
Journal title
The American statistician → ACNP
ISSN journal
00031305
Volume
51
Issue
1
Year of publication
1997
Pages
55 - 58
Database
ISI
SICI code
0003-1305(1997)51:1<55:CSOSBT>2.0.ZU;2-Q
Abstract
In a linear model under normality it is shown that the error sum of sq uares is characterized by its distribution. Two proofs are presented, one using the almost-sure uniqueness of uniformly minimum variance unb iased estimators and the other using linear algebra. Two illustrations of how this characterization can be used are given.