A dimensional reduction method for ordinal three-way contingency table

Luigi D'Ambra; Biagio Simonetti; Eric J. Beh

A dimensional reduction method for ordinal three-way contingency table

Luigi D'Ambra, Biagio Simonetti, Eric J. Beh

Research output: Chapter in Book / Conference Paper › Conference Paper

Abstract

![CDATA[For the study of association in three-way, and more generally multiway, contingency tables the literature offers a large number of techniques that can be considered. When there is an asymmetric dependence structure between the variables the Marcotorchino index [Mar84] (as apposed to the Pearson chi-squared statistic) can be used to measure the strength of their association. When the variables have an ordinal structure, this information is often not take into account. In this paper we introduce a partition of the Marcotorchino index for three ordered categorical variables using a special class of orthogonal polynomials. A graphical procedure is also considered to obtain a visual summary of the asymmetrical relationship between the variables.]]

Original language	English
Title of host publication	COMPSTAT : proceedings in computational statistics : 17th symposium held in Rome, Italy, 2006
Publisher	Physica-Verlag
Number of pages	13
ISBN (Electronic)	3790817082
ISBN (Print)	9783790817089
Publication status	Published - 2006
Event	Symposium on Computational Statistics - Duration: 1 Jan 2006 → …

Conference

Conference	Symposium on Computational Statistics
Period	1/01/06 → …

Keywords

3-way contingency table
ordinal variables
orthogonal polynomials

Cite this

@inproceedings{668d099240a3470fae4a757aa8b05c19,

title = "A dimensional reduction method for ordinal three-way contingency table",

abstract = "![CDATA[For the study of association in three-way, and more generally multiway, contingency tables the literature offers a large number of techniques that can be considered. When there is an asymmetric dependence structure between the variables the Marcotorchino index [Mar84] (as apposed to the Pearson chi-squared statistic) can be used to measure the strength of their association. When the variables have an ordinal structure, this information is often not take into account. In this paper we introduce a partition of the Marcotorchino index for three ordered categorical variables using a special class of orthogonal polynomials. A graphical procedure is also considered to obtain a visual summary of the asymmetrical relationship between the variables.]]",

keywords = "3-way contingency table, ordinal variables, orthogonal polynomials",

author = "Luigi D'Ambra and Biagio Simonetti and Beh, {Eric J.}",

year = "2006",

language = "English",

isbn = "9783790817089",

booktitle = "COMPSTAT : proceedings in computational statistics : 17th symposium held in Rome, Italy, 2006",

publisher = "Physica-Verlag",

note = "Symposium on Computational Statistics ; Conference date: 01-01-2006",

}

TY - GEN

T1 - A dimensional reduction method for ordinal three-way contingency table

AU - D'Ambra, Luigi

AU - Simonetti, Biagio

AU - Beh, Eric J.

PY - 2006

Y1 - 2006

N2 - ![CDATA[For the study of association in three-way, and more generally multiway, contingency tables the literature offers a large number of techniques that can be considered. When there is an asymmetric dependence structure between the variables the Marcotorchino index [Mar84] (as apposed to the Pearson chi-squared statistic) can be used to measure the strength of their association. When the variables have an ordinal structure, this information is often not take into account. In this paper we introduce a partition of the Marcotorchino index for three ordered categorical variables using a special class of orthogonal polynomials. A graphical procedure is also considered to obtain a visual summary of the asymmetrical relationship between the variables.]]

AB - ![CDATA[For the study of association in three-way, and more generally multiway, contingency tables the literature offers a large number of techniques that can be considered. When there is an asymmetric dependence structure between the variables the Marcotorchino index [Mar84] (as apposed to the Pearson chi-squared statistic) can be used to measure the strength of their association. When the variables have an ordinal structure, this information is often not take into account. In this paper we introduce a partition of the Marcotorchino index for three ordered categorical variables using a special class of orthogonal polynomials. A graphical procedure is also considered to obtain a visual summary of the asymmetrical relationship between the variables.]]

KW - 3-way contingency table

KW - ordinal variables

KW - orthogonal polynomials

UR - http://handle.uws.edu.au:8081/1959.7/11127

M3 - Conference Paper

SN - 9783790817089

BT - COMPSTAT : proceedings in computational statistics : 17th symposium held in Rome, Italy, 2006

PB - Physica-Verlag

T2 - Symposium on Computational Statistics

Y2 - 1 January 2006

ER -

A dimensional reduction method for ordinal three-way contingency table

Abstract

Conference

Keywords

Other files and links

Fingerprint

Cite this