Significance analysis for pairwise variable selection in classification

DOI10.4310/SII.2014.V7.N2.A11MaRDI QIDQ1748876zbMATH OpenFDO

Authors Xingye Qiao, Yufeng Liu, J. S. Marron

Publication date 14 May 2018

Published in Statistics and Its Interface (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1402.4459

zbMATH Keywords

permutation testjoint effectbivariate effectmarginal screening

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Applications of statistics to biology and medical sciences; meta analysis (62P10)

Abstract: The goal of this article is to select important variables that can distinguish one class of data from another. A marginal variable selection method ranks the marginal effects for classification of individual variables, and is a useful and efficient approach for variable selection. Our focus here is to consider the bivariate effect, in addition to the marginal effect. In particular, we are interested in those pairs of variables that can lead to accurate classification predictions when they are viewed jointly. To accomplish this, we propose a permutation test called Significance test of Joint Effect (SigJEff). In the absence of joint effect in the data, SigJEff is similar or equivalent to many marginal methods. However, when joint effects exist, our method can significantly boost the performance of variable selection. Such joint effects can help to provide additional, and sometimes dominating, advantage for classification. We illustrate and validate our approach using both simulated example and a real glioblastoma multiforme data set, which provide promising results.

Recommendations

• Embedded variable selection method using signomial classification
• Variable selection for clustering and classification
• Simultaneous feature selection and extraction using feature significance
• A criterion for variable selection in multiple discriminant analysis
• Variable selection for classification and regression in large \(p\), small \(n\) problems
• scientific article; zbMATH DE number 3974096
• Variable selection for binary classification in large dimensions: comparisons and application to microarray data
• Significance analysis of high-dimensional, low-sample size partially labeled data
• A penalized criterion for variable selection in classification

Cited in

(3)

• Why significant variables aren't automatically good predictors
• Embedded variable selection method using signomial classification
• scientific article; zbMATH DE number 4005366 (Why is no real title available?)

This page was built for publication: Significance analysis for pairwise variable selection in classification

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1748876)