On Generalized Wishart Distributions - I: Likelihood Ratio Test for Homogeneity of Covariance Matrices

被引：7

作者：

Caro-Lopera F.J. ^{[1
]}

González-Farías G. ^{[2
]}

Balakrishnan N. ^{[3
,4
]}

机构：

[1] Departamento de Ciencias Básicas, Universidad de Medellín, Medellín

[2] Centro de Investigación en Matemáticas, Monterrey

[3] Department of Mathematics and Statistics, McMaster University, Hamilton, ON

[4] Department of Statistics, King AbdulAziz University, Jeddah

来源：

Sankhya A | 2014年 / 76卷 / 2期

基金：

加拿大自然科学与工程研究理事会;

关键词：

Likelihood ratio statistic; elliptical models; zonal polynomials; Mellin transform; H-function; Primary 62E15, 60E05, 62H99; Secondary 33D70;

D O I：

10.1007/s13171-013-0047-7

中图分类号：

学科分类号：

摘要：

In this paper, we define and discuss a class of generalized Wishart distributions under elliptical models. We derive the non-central moments of the likelihood ratio statistic for testing the equality of two covariance matrices under elliptical models for the corresponding matrices. Known classical expressions for the Gaussian model are then deduced from these general results. Finally, the exact distribution of the Wilks’ statistic under a specific distribution, including the Gaussian distribution as a particular member, is derived. © 2014, Indian Statistical Institute.

引用

页码：179 / 194

页数：15