Unbiased modified likelihood ratio tests for simple and double separability of a variance–covariance structure

We present modified likelihood ratio tests (LRTs) for simple and double separability of a variance–covariance structure, unbiased in finite samples. The modification is a penalty-based homothetic transformation of the LRT statistic. Optimal penalties, depending on the mean model, contain novel infor...

Full description

Saved in:
Bibliographic Details
Published inStatistics & probability letters Vol. 83; no. 2; pp. 631 - 636
Main Authors Manceur, A.M., Dutilleul, P.
Format Journal Article
LanguageEnglish
Published Elsevier B.V 01.02.2013
Subjects
Bias in likelihood ratio testing
Modified test statistic
Multi-dimensional normal distribution models
Penalty factor
Separable variance–covariance structures
Penalty factor
Multi-dimensional normal distribution models
Separable variance–covariance structures
Modified test statistic
Bias in likelihood ratio testing
Online AccessGet full text

Cover

More Information
Summary:We present modified likelihood ratio tests (LRTs) for simple and double separability of a variance–covariance structure, unbiased in finite samples. The modification is a penalty-based homothetic transformation of the LRT statistic. Optimal penalties, depending on the mean model, contain novel information.
ISSN:0167-7152
1879-2103
DOI:10.1016/j.spl.2012.10.020