Post-selection inference via algorithmic stability

DOI10.1214/23-AOS2303arXiv2011.09462OpenAlexW3102258566MaRDI QIDQ6183754FDOQ6183754

Authors: Tijana Zrnic, Michael Jordan 

Publication date: 4 January 2024

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

Abstract: When the target of statistical inference is chosen in a data-driven manner, the guarantees provided by classical theories vanish. We propose a solution to the problem of inference after selection by building on the framework of algorithmic stability, in particular its branch with origins in the field of differential privacy. Stability is achieved via randomization of selection and it serves as a quantitative measure that is sufficient to obtain non-trivial post-selection corrections for classical confidence intervals. Importantly, the underpinnings of algorithmic stability translate directly into computational efficiency -- our method computes simple corrections for selective inference without recourse to Markov chain Monte Carlo sampling.

Full work available at URL: https://arxiv.org/abs/2011.09462

zbMATH Keywords

model selectionlinear regressionstabilityselective inferencedifferential privacypost-selection inference

Mathematics Subject Classification ID

Linear regression; mixed models (62J05) Statistical ranking and selection procedures (62F07) Paired and multiple comparisons; multiple testing (62J15)

Cites Work

• 10.1162/153244303322753616
• Title not available (Why is that?)
• Sure Independence Screening for Ultrahigh Dimensional Feature Space
• Uniformly valid confidence intervals post-model-selection
• A significance test for the lasso
• Bootstrapping and sample splitting for high-dimensional, assumption-lean inference
• Exact post-selection inference, with application to the Lasso
• Valid post-selection inference
• Predictive inference with the jackknife+
• 10.1162/153244302760200704
• Selective inference with a randomized response
• Statistical learning and selective inference
• Simultaneous and selective inference: Current successes and future challenges
• Theory of Cryptography
• Conditional predictive inference for stable algorithms
• What can we learn privately?
• Preserving Statistical Validity in Adaptive Data Analysis
• Algorithmic stability for adaptive data analysis
• The Algorithmic Foundations of Differential Privacy
• Title not available (Why is that?)
• Bounds on the sample complexity for private learning and private data release
• Valid post-selection inference in model-free linear regression
• Inferactive data analysis
• Models as approximations. I. Consequences illustrated with linear regression
• The reusable holdout: Preserving validity in adaptive data analysis
• On the Length of Post-Model-Selection Confidence Intervals Conditional on Polyhedral Constraints
• Splitting strategies for post-selection inference

Cited In (3)

• Locally simultaneous inference
• Integrative methods for post-selection inference under convex constraints
• A (tight) upper bound for the length of confidence intervals with conditional coverage

This page was built for publication: Post-selection inference via algorithmic stability

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6183754)