^2-confidence sets in high-dimensional regression

DOI10.1007/978-3-319-27099-9_13MaRDI QIDQ2956755zbMATH OpenOpenAlexFDO

Authors Sara Van De Geer, Benjamin Stucky

Publication date 19 January 2017

Published in Statistical Analysis for High-Dimensional Data (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1502.07131

Mathematics Subject Classification ID

Parametric tolerance and confidence regions (62F25) Linear regression; mixed models (62J05) Ridge regression; shrinkage estimators (Lasso) (62J07)

Abstract: We study a high-dimensional regression model. Aim is to construct a confidence set for a given group of regression coefficients, treating all other regression coefficients as nuisance parameters. We apply a one-step procedure with the square-root Lasso as initial estimator and a multivariate square-root Lasso for constructing a surrogate Fisher information matrix. The multivariate square-root Lasso is based on nuclear norm loss with $el{l}_1$-penalty. We show that this procedure leads to an asymptotically $ch{i}^2$-distributed pivot, with a remainder term depending only on the $el{l}_1$-error of the initial estimator. We show that under $el{l}_1$-sparsity conditions on the regression coefficients $$ the square-root Lasso produces to a consistent estimator of the noise variance and we establish sharp oracle inequalities which show that the remainder term is small under further sparsity conditions on $$ and compatibility conditions on the design.

Recommendations

• Confidence sets in sparse regression
• Honest Confidence Sets for High-Dimensional Regression by Projection and Shrinkage
• Confidence Intervals and Hypothesis Testing for High-Dimensional Regression
• Uniformly valid confidence sets based on the Lasso
• Confidence sets based on thresholding estimators in high-dimensional Gaussian regression models

Cited in

(15)

• Rejoinder on: ``High-dimensional simultaneous inference with the bootstrap
• Uniformly valid confidence sets based on the Lasso
• Multicarving for high-dimensional post-selection inference
• Scale calibration for high-dimensional robust regression
• Confidence sets in sparse regression
• Confidence Intervals and Hypothesis Testing for High-Dimensional Regression
• Conservative confidence intervals on multiple correlation coefficient for high-dimensional elliptical data using random projection methodology
• Honest Confidence Sets for High-Dimensional Regression by Projection and Shrinkage
• Smooth over-parameterized solvers for non-smooth structured optimization
• The EAS approach to variable selection for multivariate response data in high-dimensional settings
• Noise covariance estimation in multi-task high-dimensional linear models
• Group inference in high dimensions with applications to hierarchical testing
• Group inference of high-dimensional single-index models
• Estimation of the error structure in multivariate response linear regression models
• Statistical inference in sparse high-dimensional additive models

This page was built for publication: \(\chi^{2}\)-confidence sets in high-dimensional regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2956755)