TITLE

Nonlinear Robust Regressions Based on α-Regression Quantile, Least Median of Squares and Least Trimmed Squares Using Genetic Algorithms

AUTHOR(S)
Wibowo, Antoni Antoni; Desa, Mohamad Ishak
PUB. DATE
December 2011
SOURCE
Journal of Computing;Dec2011, Vol. 3 Issue 12, p64
SOURCE TYPE
Academic Journal
DOC. TYPE
Article
ABSTRACT
Kernel principal component regression (KPCR) can be effectively used for nonlinear system by mapping an original input space into a higher-dimensional feature space. However, KPCR can be inappropriate to be used when our data contain outliers. Under this circumstance, we propose several nonlinear robust techniques using the hybridization of KPCA, É‘- regression quantile, Least Median of Squares (LMS), Least Trimmed Squares (LTS), and genetic algorithms for handling the effects of outliers on regression models. KPCA is performed to construct nonlinearity while É‘-regression quantile, LMS and LTS are used to perform robustness of regressions. The genetic algorithms are used to estimate the regression coefficients of É‘- regression quantile, LMS and LTS methods. The performances of the proposed methods are compared to KPCR and give better results than KPCR. .
ACCESSION #
76287007

 

Share

Read the Article

Courtesy of THE LIBRARY OF VIRGINIA

Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics