A Modified Principal Component Technique Based on the LASSO

In many multivariate statistical techniques, a set of linear functions of the original p variables is produced. One of the more difficult aspects of these techniques is the interpretation of the linear functions, as these functions usually have nonzero coefficients on all p variables. A common appro...

Full description

Saved in:
Bibliographic Details
Published in:Journal of computational and graphical statistics 2003-09, Vol.12 (3), p.531-547
Main Authors: Jolliffe, Ian T, Trendafilov, Nickolay T, Uddin, Mudassir
Format: Article
Language:English
Subjects:
Coefficients
Correlations
Covariance matrices
Datasets
Eigenvectors
Interpretation
Linear transformations
Matrices
Principal component analysis
Principal components analysis
Simplification
Statistical variance
Zero
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In many multivariate statistical techniques, a set of linear functions of the original p variables is produced. One of the more difficult aspects of these techniques is the interpretation of the linear functions, as these functions usually have nonzero coefficients on all p variables. A common approach is to effectively ignore (treat as zero) any coefficients less than some threshold value, so that the function becomes simple and the interpretation becomes easier for the users. Such a procedure can be misleading. There are alternatives to principal component analysis which restrict the coefficients to a smaller number of possible values in the derivation of the linear functions, or replace the principal components by "principal variables." This article introduces a new technique, borrowing an idea proposed by Tibshirani in the context of multiple regression where similar problems arise in interpreting regression equations. This approach is the so-called LASSO, the "least absolute shrinkage and selection operator," in which a bound is introduced on the sum of the absolute values of the coefficients, and in which some coefficients consequently become zero. We explore some of the properties of the new technique, both theoretically and using simulation studies, and apply it to an example.
ISSN:1061-8600
1537-2715
DOI:10.1198/1061860032148