|
Activity Number:
|
377
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #309939 |
|
Title:
|
Likelihood Ratio Tests of Correlated Multivariate Samples
|
|
Author(s):
|
Erning Li*+ and Johan Lim
|
|
Companies:
|
Texas A&M University and Texas A&M University
|
|
Address:
|
Department of Statistics, College Station, TX, 77843-3143,
|
|
Keywords:
|
Correlated samples ; Empirical rejection probability ; Equality of mean vectors ; Homogeneity of covariance matrices ; Multivariate analysis ; Resampling
|
|
Abstract:
|
We develop methods to compare multiple multivariate samples which may be correlated. The methods are new in the context that no assumption is made about the correlations among the samples to be compared. The samples are assumed to be multivariate normally distributed and balanced. Three types of null hypothesis are considered: equality of mean vectors, homogeneity of covariance matrices, and equality of both mean vectors and covariance matrices. Finite sample and asymptotic properties of likelihood ratio tests are derived. Following the theoretical findings, we propose a resampling procedure for the implementation of the likelihood ratio tests in which no restrictive assumption is imposed on the structures of the covariance matrices. The performance of the testing procedures is investigated using simulations and via a dental health study.
|
