Publication
Title
Robust principal component analysis based on pairwise correlation estimators
Author
Abstract
Principal component analysis tries to explain and simplify the structure of multivariate data. For standardized variables, these principal components correspond to the eigenvectors of their correlation matrix. To obtain a robust principal components analysis, we estimate this correlation matrix componentwise by using robust pairwise correlation estimates. We show that the approach based on pairwise correlation estimators does not need a majority of outlier-free observations which becomes very useful for high dimensional problems. We further demonstrate that the "bivariate trimming" method especially works well in this setting.
Language
English
Source (journal)
COMPSTAT'2010: 19TH INTERNATIONAL CONFERENCE ON COMPUTATIONAL STATISTICS
Source (book)
19th International Conference on Computational Statistics, (COMPSTAT'2010), AUG 22-27, 2010, Paris, FRANCE
Publication
Heidelberg : Physica-verlag gmbh & co, 2010
ISBN
978-3-7908-2603-6
Volume/pages
(2010), p. 573-580
ISI
000395720500059
Full text (Publisher's DOI)
Full text (publisher's version - intranet only)
UAntwerpen
Faculty/Department
Research group
Publication type
Subject
Affiliation
Publications with a UAntwerp address
External links
Web of Science
Record
Identification
Creation 02.08.2018
Last edited 15.07.2021
To cite this reference