Distributional results for thresholding estimators in high-dimensional Gaussian regression models

DOI10.1214/11-EJS659zbMATH Open1271.62149arXiv1106.6002OpenAlexW3104845868MaRDI QIDQ1952253FDOQ1952253

Authors: Benedikt M. Pötscher, Ulrike Schneider 

Publication date: 28 May 2013

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Abstract: We study the distribution of hard-, soft-, and adaptive soft-thresholding estimators within a linear regression model where the number of parameters k can depend on sample size n and may diverge with n. In addition to the case of known error-variance, we define and study versions of the estimators when the error-variance is unknown. We derive the finite-sample distribution of each estimator and study its behavior in the large-sample limit, also investigating the effects of having to estimate the variance when the degrees of freedom n-k does not tend to infinity or tends to infinity very slowly. Our analysis encompasses both the case where the estimators are tuned to perform consistent model selection and the case where the estimators are tuned to perform conservative model selection. Furthermore, we discuss consistency, uniform consistency and derive the uniform convergence rate under either type of tuning.

Full work available at URL: https://arxiv.org/abs/1106.6002

Recommendations

• Confidence sets based on thresholding estimators in high-dimensional Gaussian regression models
• On the distribution of penalized maximum likelihood estimators: the LASSO, SCAD, and thresholding
• Thresholding least-squares inference in high-dimensional regression models
• Hard thresholding regression
• Estimation in high-dimensional linear models with deterministic design matrices

zbMATH Keywords

variable selectionpenalized maximum likelihoodvariance estimationoracle propertyasymptotic distributionsadaptive Lassouniform convergence ratehigh-dimensional modelfinite-sample distributions

Mathematics Subject Classification ID

Exact distribution theory in statistics (62E15) Point estimation (62F10) Linear regression; mixed models (62J05) Asymptotic distribution theory in statistics (62E20)

Cites Work

• Nearly unbiased variable selection under minimax concave penalty
• The Adaptive Lasso and Its Oracle Properties
• Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
• Title not available (Why is that?)
• Title not available (Why is that?)
• Asymptotics for Lasso-type estimators.
• Sparse estimators and the oracle property, or the return of Hodges' estimator
• MODEL SELECTION AND INFERENCE: FACTS AND FICTION
• On the distribution of penalized maximum likelihood estimators: the LASSO, SCAD, and thresholding
• Nonconcave penalized likelihood with a diverging number of parameters.
• A Statistical View of Some Chemometrics Regression Tools
• On the distribution of the adaptive LASSO estimator
• Confidence sets based on penalized maximum likelihood estimators in Gaussian regression
• THE FINITE-SAMPLE DISTRIBUTION OF POST-MODEL-SELECTION ESTIMATORS AND UNIFORM VERSUS NONUNIFORM APPROXIMATIONS
• The distribution of model averaging estimators and an impossibility result regarding its estima\-tion
• Asymptotic properties of maximum likelihood estimators based on conditional specification
• Title not available (Why is that?)
• Title not available (Why is that?)
• Model selection by multiple test procedures

Cited In (14)

• Excess-risk consistency of group-hard thresholding estimator in robust estimation of Gaussian mean
• Generalized thresholding estimators for high-dimensional location parameters
• On various confidence intervals post-model-selection
• The adaptive Lasso in high-dimensional sparse heteroscedastic models
• Canonical thresholding for nonsparse high-dimensional linear regression
• Confidence sets in sparse regression
• Thresholding least-squares inference in high-dimensional regression models
• Uniformly valid confidence sets based on the Lasso
• On consistency for time series model selection
• Asymptotics for threshold regression under general conditions
• Hard thresholding regression
• Valid post-selection inference
• Confidence sets based on thresholding estimators in high-dimensional Gaussian regression models
• On the distribution of penalized maximum likelihood estimators: the LASSO, SCAD, and thresholding

This page was built for publication: Distributional results for thresholding estimators in high-dimensional Gaussian regression models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1952253)