Confidence intervals for high-dimensional linear regression: minimax rates and adaptivity

DOI10.1214/16-AOS1461zbMATH Open1371.62045arXiv1506.05539OpenAlexW1666335228MaRDI QIDQ2012200FDOQ2012200

Zi-Jian Guo, T. Tony Cai

Publication date: 28 July 2017

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

Abstract: Confidence sets play a fundamental role in statistical inference. In this paper, we consider confidence intervals for high dimensional linear regression with random design. We first establish the convergence rates of the minimax expected length for confidence intervals in the oracle setting where the sparsity parameter is given. The focus is then on the problem of adaptation to sparsity for the construction of confidence intervals. Ideally, an adaptive confidence interval should have its length automatically adjusted to the sparsity of the unknown regression vector, while maintaining a prespecified coverage probability. It is shown that such a goal is in general not attainable, except when the sparsity parameter is restricted to a small region over which the confidence intervals have the optimal length of the usual parametric rate. It is further demonstrated that the lack of adaptivity is not due to the conservativeness of the minimax framework, but is fundamentally caused by the difficulty of learning the bias accurately.

Full work available at URL: https://arxiv.org/abs/1506.05539

Recommendations

• Confidence sets in sparse regression
• Confidence intervals for low dimensional parameters in high dimensional linear models
• Exact adaptive confidence intervals for linear regression coefficients
• Confidence Intervals and Hypothesis Testing for High-Dimensional Regression
• Confidence sets based on thresholding estimators in high-dimensional Gaussian regression models

zbMATH Keywords

confidence intervalexpected lengthsparsityadaptivitycoverage probabilityhigh-dimensional linear regressionminimaxity

Mathematics Subject Classification ID

Linear regression; mixed models (62J05) Nonparametric tolerance and confidence regions (62G15) Minimax procedures in statistical decision theory (62C20)

Cited In (61)

• Adaptive estimation of high-dimensional signal-to-noise ratios
• Integrative Factor Regression and Its Inference for Multimodal Data Analysis
• Accuracy assessment for high-dimensional linear regression
• On lower bounds for the bias-variance trade-off
• Design of c-optimal experiments for high-dimensional linear models
• A unified theory of confidence regions and testing for high-dimensional estimating equations
• Empirical Bayes oracle uncertainty quantification for regression
• Projection-based Inference for High-dimensional Linear Models
• On the asymptotic variance of the debiased Lasso
• Debiasing the Lasso: optimal sample size for Gaussian designs
• Optimal sparsity testing in linear regression model
• Testability of high-dimensional linear models with nonsparse structures
• Confidence Intervals and Hypothesis Testing for High-Dimensional Regression
• Sparse Topic Modeling: Computational Efficiency, Near-Optimal Algorithms, and Statistical Inference
• Thresholding tests based on affine Lasso to achieve non-asymptotic nominal level and high power under sparse and dense alternatives in high dimension
• Dynamic linear discriminant analysis in high dimensional space
• Orthogonalized Kernel Debiased Machine Learning for Multimodal Data Analysis
• Confidence Intervals for Maximin Effects in Inhomogeneous Large-Scale Data
• Statistical Inference for High-Dimensional Generalized Linear Models With Binary Outcomes
• Multicarving for high-dimensional post-selection inference
• Linear Hypothesis Testing in Dense High-Dimensional Linear Models
• Efficient estimation of linear functionals of principal components
• Controlling False Discovery Rate Using Gaussian Mirrors
• Predictor ranking and false discovery proportion control in high-dimensional regression
• Group Inference in High Dimensions with Applications to Hierarchical Testing
• Variable selection via adaptive false negative control in linear regression
• Honest Confidence Sets for High-Dimensional Regression by Projection and Shrinkage
• Testing Endogeneity with High Dimensional Covariates
• Title not available (Why is that?)
• Title not available (Why is that?)
• Double-estimation-friendly inference for high-dimensional misspecified models
• Statistical inference via conditional Bayesian posteriors in high-dimensional linear regression
• Minimum-volume confidence sets for normal linear regression models
• Scale calibration for high-dimensional robust regression
• Some perspectives on inference in high dimensions
• Fixed Effects Testing in High-Dimensional Linear Mixed Models
• Debiasing convex regularized estimators and interval estimation in linear models
• Comments on: ``High-dimensional simultaneous inference with the bootstrap
• Model-Assisted Uniformly Honest Inference for Optimal Treatment Regimes in High Dimension
• Debiasing the debiased Lasso with bootstrap
• Inference and Estimation for Random Effects in High-Dimensional Linear Mixed Models
• De-biasing the Lasso with degrees-of-freedom adjustment
• Global and Simultaneous Hypothesis Testing for High-Dimensional Logistic Regression Models
• Robust post-selection inference of high-dimensional mean regression with heavy-tailed asymmetric or heteroskedastic errors
• Significance testing in non-sparse high-dimensional linear models
• High-Dimensional Inference for Cluster-Based Graphical Models
• Doubly debiased Lasso: high-dimensional inference under hidden confounding
• Adaptive estimation of the sparsity in the Gaussian vector model
• Semiparametric efficiency bounds for high-dimensional models
• Inference for High-Dimensional Linear Mixed-Effects Models: A Quasi-Likelihood Approach
• Stochastic continuum-armed bandits with additive models: minimax regrets and adaptive algorithm
• IFAA: Robust Association Identification and Inference for Absolute Abundance in Microbiome Analyses
• Estimation of Linear Functionals in High-Dimensional Linear Models: From Sparsity to Nonsparsity
• Test of Significance for High-Dimensional Thresholds with Application to Individualized Minimal Clinically Important Difference
• Online Debiasing for Adaptively Collected High-Dimensional Data With Applications to Time Series Analysis
• LIC criterion for optimal subset selection in distributed interval estimation
• Inference for high-dimensional linear expectile regression with de-biasing method
• Statistical Inference for Maximin Effects: Identifying Stable Associations across Multiple Studies
• Transfer Learning with Large-Scale Quantile Regression
• Inference for heteroskedastic PCA with missing data
• Estimation and Inference for High-Dimensional Generalized Linear Models with Knowledge Transfer

This page was built for publication: Confidence intervals for high-dimensional linear regression: minimax rates and adaptivity

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2012200)