On various confidence intervals post-model-selection

DOI10.1214/14-STS507zbMATH Open1332.62154arXiv1401.2267OpenAlexW1933929712MaRDI QIDQ254446FDOQ254446

Authors: Hannes Leeb, Benedikt M. Pötscher, Karl Ewald 

Publication date: 8 March 2016

Published in: Statistical Science (Search for Journal in Brave)

Abstract: We compare several confidence intervals after model selection in the setting recently studied by Berk et al. [Ann. Statist. 41 (2013) 802-837], where the goal is to cover not the true parameter but a certain nonstandard quantity of interest that depends on the selected model. In particular, we compare the PoSI-intervals that are proposed in that reference with the "naive" confidence interval, which is constructed as if the selected model were correct and fixed a priori (thus ignoring the presence of model selection). Overall, we find that the actual coverage probabilities of all these intervals deviate only moderately from the desired nominal coverage probability. This finding is in stark contrast to several papers in the existing literature, where the goal is to cover the true parameter.

Full work available at URL: https://arxiv.org/abs/1401.2267

Recommendations

• Valid confidence intervals for post-model-selection predictors
• Model-averaged confidence intervals
• Uniformly valid confidence intervals post-model-selection
• Two sources of poor coverage of confidence intervals after model selection
• On the Large-Sample Minimal Coverage Probability of Confidence Intervals After Model Selection

zbMATH Keywords

Lassomodel selectionBICAICconfidence intervalsnonstandard coverage target

Mathematics Subject Classification ID

Nonparametric tolerance and confidence regions (62G15)

Cites Work

• Least angle regression. (With discussion)
• Smoothing noisy data with spline functions: Estimating the correct degree of smoothing by the method of generalized cross-validation
• Valid post-selection inference
• Hybrid and Size-Corrected Subsampling Methods
• Title not available (Why is that?)
• MODEL SELECTION AND INFERENCE: FACTS AND FICTION
• On the distribution of penalized maximum likelihood estimators: the LASSO, SCAD, and thresholding
• Confidence sets based on sparse estimators are necessarily large
• Can one estimate the conditional distribution of post-model-selection estimators?
• On the Large-Sample Minimal Coverage Probability of Confidence Intervals After Model Selection
• Evaluation and selection of models for out-of-sample prediction when the sample size is small relative to the complexity of the data-generating process
• On the distribution of the adaptive LASSO estimator
• On preliminary test and shrinkage M-estimation in linear models
• Confidence sets based on penalized maximum likelihood estimators in Gaussian regression
• Distributional results for thresholding estimators in high-dimensional Gaussian regression models
• The Conditional Level of the F-Test
• PERFORMANCE LIMITS FOR ESTIMATORS OF THE RISK OR DISTRIBUTION OF SHRINKAGE-TYPE ESTIMATORS, AND SOME GENERAL LOWER RISK-BOUND RESULTS
• CAN ONE ESTIMATE THE UNCONDITIONAL DISTRIBUTION OF POST-MODEL-SELECTION ESTIMATORS?
• Applied Regression Analysis
• THE FINITE-SAMPLE DISTRIBUTION OF POST-MODEL-SELECTION ESTIMATORS AND UNIFORM VERSUS NONUNIFORM APPROXIMATIONS
• The distribution of a linear predictor after model selection: unconditional finite-sample distributions and asymptotic approximations
• The distribution of model averaging estimators and an impossibility result regarding its estima\-tion
• The Conditional Level of Student's $t$ Test
• Note on a Conditional Property of Student's $t^1$
• Asymptotic properties of maximum likelihood estimators based on conditional specification

Cited In (24)

• Fast multiscale functional estimation in optimal EMG placement for robotic prosthesis controllers
• Uniformly valid confidence intervals post-model-selection
• Confidence intervals centred on bootstrap smoothed estimators
• Targeted Inference Involving High-Dimensional Data Using Nuisance Penalized Regression
• On the length of post-model-selection confidence intervals conditional on polyhedral constraints
• Confidence intervals for high-dimensional partially linear single-index models
• Post-model-selection inference in linear regression models: an integrated review
• Two sources of poor coverage of confidence intervals after model selection
• Upper bounds on the minimum coverage probability of confidence intervals in regression after model selection
• Optimal EMG placement for a robotic prosthesis controller with sequential, adaptive functional estimation (SAFE)
• Bootstrapping some GLM and survival regression variable selection estimators
• Confidence sets for model selection by F -testing
• Comments on “Unobservable Selection and Coefficient Stability: Theory and Evidence” and “Poorly Measured Confounders are More Useful on the Left Than on the Right”
• Valid confidence intervals for post-model-selection predictors
• In defense of the indefensible: a very naïve approach to high-dimensional inference
• Bootstrapping multiple linear regression after variable selection
• Post-model-selection prediction intervals for generalized linear models
• Survival analysis of DNA mutation motifs with penalized proportional hazards
• Simultaneous confidence bands for contrasts between several nonlinear regression curves
• ON THE COVERAGE PROBABILITY OF CONFIDENCE INTERVALS IN REGRESSION AFTER VARIABLE SELECTION
• Discussion on “Model confidence bounds for variable selection” by Yang Li, Yuetian Luo, Davide Ferrari, Xiaonan Hu, and Yichen Qin
• On the Large-Sample Minimal Coverage Probability of Confidence Intervals After Model Selection
• Statistical proof? The problem of irreproducibility
• Mixture of Linear Models Co-supervised by Deep Neural Networks

This page was built for publication: On various confidence intervals post-model-selection

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q254446)