Publication detail

Approximation of powers of some tests in one-way MANOVA type multivariate generalized linear model

HRDLIČKOVÁ, Z.

Czech title

Aproximace sil některých testu v mnohorozměrném zobecněném modelu typu jednoduché MANOVY

English title

Approximation of powers of some tests in one-way MANOVA type multivariate generalized linear model

Type

journal article - other

Language

en

Original abstract

The test of an equality of expected values of response vectors in one-way MANOVA type multivariate generalized linear models can be based on deviance or score statistic. This paper aims at approximation of the power of these tests under local alternatives. As indicated by presented simulation study, these approximations perform well over a wide range of models and sample sizes. Approximation of the power of these tests based on a noncentral chi-square distribution derived by Fahrmeir [Fahrmeir, L., 1987. Asymptotic testing theory for generalized linear models. Statistics 18 (1), 65-76] is considered for comparison in the simulation study. The approximation derived in this paper performs better for small sample sizes. The simulation study also provides comparison of the considered tests.

Czech abstract

Test shodnosti středních hodnot závislých vektorů v mnohorozměrném zobecněném lineárním modelu typu jednoduché MANOVY může byt založen na devianci nebo skórové statistice. Tento článek je zaměřen na aproximaci sil těchto testů při lokální alternativě. Jak vyplývá z uvedené simulační studie, jsou tyto aproximace vhodné pro široký okruh modelů a rozsahy výběru. V simulační studii je pro srovnání uvažována aproximace sil těchto testů založená na necentrálním chi-kvadrátu odvozena Fahrmeirem [Fahrmeir, L., 1987. Asymptotic testing theory for generalized linear models. Statistics 18 (1), 65-76] Aproximace odvozené v tomto článku jsou vhodnější pro malé rozsahy výběru. Simulační studie umožňuje srovnání uvažovaných testů.

English abstract

The test of an equality of expected values of response vectors in one-way MANOVA type multivariate generalized linear models can be based on deviance or score statistic. This paper aims at approximation of the power of these tests under local alternatives. As indicated by presented simulation study, these approximations perform well over a wide range of models and sample sizes. Approximation of the power of these tests based on a noncentral chi-square distribution derived by Fahrmeir [Fahrmeir, L., 1987. Asymptotic testing theory for generalized linear models. Statistics 18 (1), 65-76] is considered for comparison in the simulation study. The approximation derived in this paper performs better for small sample sizes. The simulation study also provides comparison of the considered tests.

Keywords in Czech

Mnohorozměrný zobecněný lineární model; Deviance; Skórová statistika; Síla; Jednoduchá MANOVA; Jednoduchá ANOVA

Keywords in English

Multivariate generalized linear model; Deviance; Score statistic; Power One-way MANOVA; One-way ANOVA

RIV year

2008

Released

16.04.2008

Publisher

Elsevier

Location

Amsterdam

ISSN

0167-9473

Journal

COMPUTATIONAL STATISTICS & DATA ANALYSIS

Volume

52

Number

8

Pages from–to

4059–4075

Pages count

17

BIBTEX


@article{BUT45042,
  author="Zuzana {Hübnerová},
  title="Approximation of powers of some tests in one-way MANOVA type multivariate generalized linear model",
  journal="COMPUTATIONAL STATISTICS & DATA ANALYSIS",
  year="2008",
  volume="52",
  number="8",
  month="April",
  pages="4059--4075",
  publisher="Elsevier",
  address="Amsterdam",
  issn="0167-9473"
}