[RT13] n extension of Non-Metric approach to inwards directed PLS Path Models

Atelier, Poster ou Démonstration dans une Conférence Internationale : SIS2014 - 47th Scientific Meeting of the Italian Statistical Society, June 2013, pp.xx, Cagliari, Italy,

Auteurs: G. Russolillo , L. Trinchera

Mots clés: Optimal Scaling, Non-Metric Variables, PLS Path Modeling

Résumé: Partial Least Squares Path Modeling (PLS-PM) aims at modeling multivariate relations among several blocks of variables observed on the same units. Since its first presentation in 1975 by Herman Wold, several authors have developed specific techniques related to the PLS-PM and investigated its mathematical, numerical and statistical properties. In recent years researchers mainly focused on the optimization criteria behind the PLS-PM algorithm and on modifications of such criteria in order to handle variables independently of the measurement scale on which they have been observed. In this paper we extend the Non-Metric approach for the analysis of variables measured on different scales presented by Russolillo in 2012 to inwards directed PLS Path Models.

BibTeX

@inproceedings {
	RT13,
	title	=	"{n extension of Non-Metric approach to inwards directed PLS Path Models}",
	author	=	" G. Russolillo and L. Trinchera ",
	booktitle	=	"{SIS2014 - 47th Scientific Meeting of the Italian Statistical Society}",
	year	=	2013,
	month	=	"June",
	pages	=	"xx",
	address	=	"Cagliari, Italy",
}