Simultaneous Detection of Shift in Means and Variances
作者:
BernardD. Flury,
DaanG. Nel,
Inet Pienaar,
期刊:
Journal of the American Statistical Association
(Taylor Available online 1995)
卷期:
Volume 90,
issue 432
页码: 1474-1481
ISSN:0162-1459
年代: 1995
DOI:10.1080/01621459.1995.10476654
出版商: Taylor & Francis Group
关键词: Allometry;Likelihood estimation;Model selection;Principal component;Randomization test;T2test
数据来源: Taylor
摘要:
Suppose thatXis ap-variate random vector, measured in two populations, with mean vector μjand covariance matrix ψjin populationj,j= 1, 2. In this article we study normal theory estimation of the parameters under the following constraints. It is assumed that (a) the covariance matrices havepidentical eigenvectors, (b) in the coordinate system given by the common eigenvectors, onlyq<pof the means are different between groups, and (c) only theqeigenvalues associated with the sameqcommon eigenvectors are different between groups. There are two main areas of application of these models: (a) in morphometric studies, particularly if size differences between groups are to be removed from the analysis, and (b) in testing for equality of mean vectors and covariance matrices, especially in situations where the number of variablespis large. For applications of type (b), we suggest a randomization test and compare its performance to theT2test and to the likelihood ratio test for equality of both populations.
点击下载:
PDF (727KB)
返 回