Piecewise-linear approximation for feature subset selection in a sequential logit model

DOI10.15807/JORSJ.60.1zbMATH Open1371.90068arXiv1510.05417OpenAlexW2963943759MaRDI QIDQ5267563FDOQ5267563

Yuichi Takano, Toshiki Sato, Ryuhei Miyashiro

Publication date: 13 June 2017

Published in: Journal of the Operations Research Society of Japan (Search for Journal in Brave)

Abstract: This paper concerns a method of selecting a subset of features for a sequential logit model. Tanaka and Nakagawa (2014) proposed a mixed integer quadratic optimization formulation for solving the problem based on a quadratic approximation of the logistic loss function. However, since there is a significant gap between the logistic loss function and its quadratic approximation, their formulation may fail to find a good subset of features. To overcome this drawback, we apply a piecewise-linear approximation to the logistic loss function. Accordingly, we frame the feature subset selection problem of minimizing an information criterion as a mixed integer linear optimization problem. The computational results demonstrate that our piecewise-linear approximation approach found a better subset of features than the quadratic approximation approach.

Full work available at URL: https://arxiv.org/abs/1510.05417

Recommendations

• Feature subset selection for logistic regression via mixed integer optimization
• An effective procedure for feature subset selection in logistic regression based on information criteria
• Variable selection in logistic regression models through the application of exact mathematical programming
• Application of a mixed integer nonlinear programming approach to variable selection in logistic regression
• Best Subsets Logistic Regression

zbMATH Keywords

optimizationstatisticsinformation criterionfeature subset selectionpiecewise-linear approximationsequential logit model

Mathematics Subject Classification ID

Quadratic programming (90C20) Management decision making, including multiple objectives (90B50) Mixed integer programming (90C11)

Cited In (7)

• Title not available (Why is that?)
• Feature subset selection for logistic regression via mixed integer optimization
• Learning Optimized Risk Scores
• An effective procedure for feature subset selection in logistic regression based on information criteria
• Mixed integer quadratic optimization formulations for eliminating multicollinearity based on variance inflation factor
• Investigating consumers' store-choice behavior via hierarchical variable selection
• Best subset selection via cross-validation criterion

This page was built for publication: Piecewise-linear approximation for feature subset selection in a sequential logit model

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5267563)