A METHOD FOR SELECTION OF VARIABLES IN THE PRINCIPAL COMPONENTS REGRESSION

Fernando FERRARI[1]

Antonio Francisco IEMMA[2]

§    ABSTRACT: In this paper we present a method for selection of variables in the principal components regression in the presence of multicollinearity. This method removes independent variables of the regression model taking in consideration the magnitude of the eigenvalues of the matrix X*’X*, where X* = [Y, X], and the prediction efficacy of independent variables on the response variable.

§    KEYWORDS: Multiple linear regression; multicollinearity; principal components.

 



[1]Departamento de Análise Numérica e Estatística - Instituto de Biociências , Letras e Ciências Exatas – UNESP – 15055 – São José do Rio Preto – SP.

[2]Departamento de Matemática - Escola Superior de Agricultura " Luiz Vaz de Queiróz" - ESALQ - USP - 13418-260 - Piracicaba - SP - Brasil.