ScholarLens Back to search
2012•Journal of Statistical Computation and SimulationRequires access

Pairwise testing of group mean vectors in MANOVA with small samples

Philip H. Ramsey, Philip H. Ramsey, Patricia P. Ramsey, Patricia P. Ramsey

Open publisher page 2 citations

Abstract

Several, multivariate, pairwise, multiple comparison procedures are proposed as follow-ups for a significant multivariate analysis of variance. The Peritz procedure is generalized from univariate to several multivariate applications. Procedures are evaluated using overall power, any-pair power and all-pairs power applied to mean vectors with common sample sizes of 4, 5, and 9. Monte Carlo simulation demonstrated greater power than previously proposed univariate procedures in many conditions especially for all-pairs power. The multivariate Peritz procedure based on the Lawley–Hotelling trace was found to be most powerful in many conditions.

About this research paper

What this paper is about

Several, multivariate, pairwise, multiple comparison procedures are proposed as follow-ups for a significant multivariate analysis of variance. The Peritz procedure is generalized from univariate to several multivariate applications. Procedures are evaluated using overall power, any-pair power and all-pairs power applied to mean vectors with common sample sizes of 4, 5, and 9. Monte Carlo simulation demonstrated greater power than previously proposed univariate procedures in many conditions especially for all-pairs power. The multivariate Peritz procedure based on the Lawley–Hotelling trace was found to be most powerful in many conditions.

Why it matters

OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Several, multivariate, pairwise, multiple comparison procedures are proposed as follow-ups for a significant multivariate analysis of variance. The Peritz procedure is generalized from univariate to several multivariate applications. Procedures are evaluated using overall power, any-pair power and all-pairs power applied to mean vectors with common sample sizes of 4, 5, and 9. Monte Carlo simulation demonstrated greater power than previously proposed univariate procedures in many conditions especially for all-pairs power. The multivariate Peritz procedure based on the Lawley–Hotelling trace was found to be most powerful in many conditions.

Key concepts: Mathematics, Multivariate analysis of variance, Pairwise comparison, Statistics, Group (periodic table), Combinatorics, Chemistry, Organic chemistry

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
Pairwise testing of group mean vectors in MANOVA with small samples — Research Paper | ScholarLens