Exact post-selection inference, with application to the lasso

Abstract

We develop a general approach to valid inference after model selection. At the core of our framework is a result that characterizes the distribution of a post-selection estimator conditioned on the selection event. We specialize the approach to model selection by the lasso to form valid confidence intervals for the selected coefficients and test whether all relevant variables have been included in the model.

Published at http://dx.doi.org/10.1214/15-AOS1371 in the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (1)

• Can one estimate the conditional distribution of post-model-selection estimators?

Cited by in corpus (50)

• Least Squares Approximation for a Distributed System
• Network classification with applications to brain connectomics
• Post hoc inference via joint family-wise error rate control
• Log-ratio Lasso: Scalable, Sparse Estimation for Log-ratio Models
• Robust Empirical Bayes Confidence Intervals
• Interpreting Black Box Models via Hypothesis Testing
• Modern approaches for evaluating treatment effect heterogeneity from clinical trials and observational data
• Estimating the error variance in a high-dimensional linear model
• An Overview of Healthcare Data Analytics With Applications to the COVID-19 Pandemic
• Testing the simplifying assumption in high-dimensional vine copulas
• High-dimensional empirical likelihood inference
• Statistical Methods for Replicability Assessment
• On the Post Selection Inference constant under Restricted Isometry Properties
• Evaluating the Effectiveness of Personalized Medicine with Software
• Generalized Data Thinning Using Sufficient Statistics
• On the hierarchical Bayesian modelling of frequency response functions
• Post-selection estimation and testing following aggregated association tests
• On the Distribution, Model Selection Properties and Uniqueness of the Lasso Estimator in Low and High Dimensions
• Uniformly Valid Confidence Sets Based on the Lasso
• Selective Inference for Testing Trees and Edges in Phylogenetics
• Optimal post-selection inference for sparse signals: a nonparametric empirical-Bayes approach
• Scalable Sparse Cox's Regression for Large-Scale Survival Data via Broken Adaptive Ridge
• The costs and benefits of uniformly valid causal inference with high-dimensional nuisance parameters
• Low-Complexity Methods for Estimation After Parameter Selection
• An efficient and robust approach to Mendelian randomization with measured pleiotropic effects in a high-dimensional setting
• Discovering Significant Topics from Legal Decisions with Selective Inference
• Iterative Hard Thresholding for Model Selection in Genome-Wide Association Studies
• Confidence intervals centered on bootstrap smoothed estimators
• Selective conformal inference with false coverage-statement rate control
• On Selecting and Conditioning in Multiple Testing and Selective Inference
• Proximal MCMC for Bayesian Inference of Constrained and Regularized Estimation
• Non-Bayesian Post-Model-Selection Estimation as Estimation Under Model Misspecification
• Two-Stage Robust and Sparse Distributed Statistical Inference for Large-Scale Data
• Uncertainty Quantification for High Dimensional Sparse Nonparametric Additive Models
• ${\ell }_1$-penalized Multinomial Regression: Estimation, inference, and prediction, with an application to risk factor identification for different dementia subtypes
• Inference on the proportion of variance explained in principal component analysis
• A Metropolized adaptive subspace algorithm for high-dimensional Bayesian variable selection
• Estimator Augmentation with Applications in High-Dimensional Group Inference
• Post hoc false discovery proportion inference under a Hidden Markov Model
• Uniformly Valid Inference Based on the Lasso in Linear Mixed Models
• Selective inference after convex clustering with ${\ell }_1$ penalization
• Sparsified Simultaneous Confidence Intervals for High-Dimensional Linear Models
• Regularization for the Approximation of 2D Set of Points via the Length of the Curve
• Variable Selection in Functional Linear Cox Model
• Post-selection Inference in Regression Models for Group Testing Data
• A Unified Framework for Pattern Recovery in Penalized and Thresholded Estimation and its Geometry
• Quantifying uncertainty in spikes estimated from calcium imaging data
• A Computationally Efficient Approach to False Discovery Rate Control and Power Maximisation via Randomisation and Mirror Statistic
• Assessment of Case Influence in the Lasso with a Case-weight Adjusted Solution Path
• High-dimensional Statistical Inference and Variable Selection Using Sufficient Dimension Association