The multi-sample block-matrix sphericity test

Carlos Manuel Agra Coelho; Filipe José Gonçalves Pereira Marques

doi:10.1063/1.3637904

The multi-sample block-matrix sphericity test

Carlos Manuel Agra Coelho, Filipe José Gonçalves Pereira Marques

DM - Departamento de Matemática

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

The multi‐sample block‐matrix sphericity test and its particular cases have wide applications in several areas. However, the practical implementation of this test has been hindered by difficulties in handling the exact distribution of the associated statistic and the non‐availability in the literature of asymptotic distributions. We use a decomposition of the null hypothesis into three null hypotheses to obtain very well‐fit and highly manageable near‐exact distributions for the likelihood ratio test statistic of this test and its particular cases. These distributions will allow for the easy computation of well‐fit near‐exact quantiles and p‐values.

Original language	Unknown
Title of host publication	AIP Conference Proceedings
Pages	1479-1482
Volume	1389
DOIs	https://doi.org/10.1063/1.3637904
Publication status	Published - 1 Jan 2011
Event	ICNAAM 2011 Numerical Analysis and Applied Mathematic - Duration: 1 Jan 2011 → …

Conference

Conference	ICNAAM 2011 Numerical Analysis and Applied Mathematic
Period	1/01/11 → …

Access to Document

10.1063/1.3637904

Cite this

@inproceedings{c704889c7f314cae896e3afe1f368ee4,

title = "The multi-sample block-matrix sphericity test",

abstract = "The multi‐sample block‐matrix sphericity test and its particular cases have wide applications in several areas. However, the practical implementation of this test has been hindered by difficulties in handling the exact distribution of the associated statistic and the non‐availability in the literature of asymptotic distributions. We use a decomposition of the null hypothesis into three null hypotheses to obtain very well‐fit and highly manageable near‐exact distributions for the likelihood ratio test statistic of this test and its particular cases. These distributions will allow for the easy computation of well‐fit near‐exact quantiles and p‐values.",

keywords = "near-exact distributions, block-matrix, sphericity test, mixtures, asymptotic distributions, multi-sample",

author = "Coelho, {Carlos Manuel Agra} and Marques, {Filipe Jos{\'e} Gon{\c c}alves Pereira}",

year = "2011",

month = jan,

day = "1",

doi = "10.1063/1.3637904",

language = "Unknown",

isbn = "978-0-7354-0956-9",

volume = "1389",

pages = "1479--1482",

booktitle = "AIP Conference Proceedings",

note = "ICNAAM 2011 Numerical Analysis and Applied Mathematic ; Conference date: 01-01-2011",

}

TY - GEN

T1 - The multi-sample block-matrix sphericity test

AU - Coelho, Carlos Manuel Agra

AU - Marques, Filipe José Gonçalves Pereira

PY - 2011/1/1

Y1 - 2011/1/1

N2 - The multi‐sample block‐matrix sphericity test and its particular cases have wide applications in several areas. However, the practical implementation of this test has been hindered by difficulties in handling the exact distribution of the associated statistic and the non‐availability in the literature of asymptotic distributions. We use a decomposition of the null hypothesis into three null hypotheses to obtain very well‐fit and highly manageable near‐exact distributions for the likelihood ratio test statistic of this test and its particular cases. These distributions will allow for the easy computation of well‐fit near‐exact quantiles and p‐values.

AB - The multi‐sample block‐matrix sphericity test and its particular cases have wide applications in several areas. However, the practical implementation of this test has been hindered by difficulties in handling the exact distribution of the associated statistic and the non‐availability in the literature of asymptotic distributions. We use a decomposition of the null hypothesis into three null hypotheses to obtain very well‐fit and highly manageable near‐exact distributions for the likelihood ratio test statistic of this test and its particular cases. These distributions will allow for the easy computation of well‐fit near‐exact quantiles and p‐values.

KW - near-exact distributions

KW - block-matrix

KW - sphericity test

KW - mixtures

KW - asymptotic distributions

KW - multi-sample

U2 - 10.1063/1.3637904

DO - 10.1063/1.3637904

M3 - Conference contribution

SN - 978-0-7354-0956-9

VL - 1389

SP - 1479

EP - 1482

BT - AIP Conference Proceedings

T2 - ICNAAM 2011 Numerical Analysis and Applied Mathematic

Y2 - 1 January 2011

ER -