Premium
Model building in multivariate additive partial least squares splines via the GCV criterion
Author(s) -
Lombardo R.,
Durand J. F.,
De Veaux R.
Publication year - 2009
Publication title -
journal of chemometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.47
H-Index - 92
eISSN - 1099-128X
pISSN - 0886-9383
DOI - 10.1002/cem.1260
Subject(s) - partial least squares regression , multivariate statistics , generalization , mathematics , nonlinear system , least squares function approximation , cross validation , partial derivative , mathematical optimization , computer science , statistics , mathematical analysis , physics , quantum mechanics , estimator
In the literature, much effort has been put into modeling dependence among variables and their interactions through nonlinear transformations of predictive variables. In this paper, we propose a nonlinear generalization of Partial Least Squares (PLS) using multivariate additive splines. We discuss the advantages and drawbacks of the proposed model, building it via the generalized cross validation criterion (GCV) criterion, and show its performance on a real dataset and on simulated datasets in comparison to other methods based on splines. Copyright © 2009 John Wiley & Sons, Ltd.