Faithful variable screening for high-dimensional convex regression

DOI10.1214/15-AOS1425MaRDI QIDQ510689zbMATH OpenOpenAlexFDO

Authors Minhua Chen, Min Xu, John Lafferty

Publication date 13 February 2017

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1411.1805

zbMATH Keywords

additive modelconvex regressionnonparametric regressionquadratic programmingvariable selection

Mathematics Subject Classification ID

Convex functions and convex programs in convex geometry (52A41) Nonparametric regression and quantile regression (62G08) Ridge regression; shrinkage estimators (Lasso) (62J07)

Abstract: We study the problem of variable selection in convex nonparametric regression. Under the assumption that the true regression function is convex and sparse, we develop a screening procedure to select a subset of variables that contains the relevant variables. Our approach is a two-stage quadratic programming method that estimates a sum of one-dimensional convex functions, followed by one-dimensional concave regression fits on the residuals. In contrast to previous methods for sparse additive models, the optimization is finite dimensional and requires no tuning parameters for smoothness. Under appropriate assumptions, we prove that the procedure is faithful in the population setting, yielding no false negatives. We give a finite sample statistical analysis, and introduce algorithms for efficiently carrying out the required quadratic programs. The approach leads to computational and statistical advantages over fitting a full model, and provides an effective, practical approach to variable screening in convex regression.

Recommendations

• Nonparametric screening under conditional strictly convex loss for ultrahigh dimensional sparse data
• High-dimensional variable selection
• Adaptive variable selection in nonparametric sparse regression
• Nonparametric independence screening in sparse ultra-high-dimensional additive models
• Sure independence screening for ultrahigh dimensional feature space. With discussion and authors' reply

Cited in

(11)

• Editorial: Special issue on ``Nonparametric inference under shape constraints
• Variable selection in convex quantile regression: \(\mathcal{L}_1\)-norm or \(\mathcal{L}_0\)-norm regularization?
• Bracketing numbers of convex and m-monotone functions on polytopes
• On the complexity of detecting convexity over a box
• Adaptation in multivariate log-concave density estimation
• On Degrees of Freedom of Projection Estimators With Applications to Multivariate Nonparametric Regression
• Sparse Convex Regression
• Non-crossing convex quantile regression
• Convex support vector regression
• Structure identification for partially linear partially concave models
• Sparse max-affine regression

This page was built for publication: Faithful variable screening for high-dimensional convex regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q510689)