Bootstrapping and sample splitting for high-dimensional, assumption-lean inference
From MaRDI portal
Publication:2284380
In recent times, there has been a crisis in the sciences because too many research results are found to lack replicability and reproducibility. Some of this crisis has been attributed to a failure of statistical methods to account for data dependent extraction and modeling that precedes statistical inference [\textit{R. Berk}et al., ``Assumption lean regression, Preprint, \url{arXiv:1806.09014}]. Valid inference after model selection is performing. An older method is sample splitting: use part of the data for model selection and the rest for inference. A new method is proposed for sample splitting combined with the bootstrap (or the normal approximation). New parameters that measure variable importance are defined and that can be inferred with greater accuracy than the usual regression coefficient.
Recommendations
- Uniform asymptotic inference and the bootstrap after model selection
- Can we trust the bootstrap in high-dimensions? The case of linear models
- A bootstrap recipe for post-model-selection inference under linear regression models
- Valid post-selection inference
- \(p\)-values for high-dimensional regression
Cites work
- \(p\)-values for high-dimensional regression
- A central limit theorem applicable to robust regression estimators
- A note on data-splitting for the evaluation of significance levels
- A significance test for the lasso
- Bootstrap consistency for quadratic forms of sample averages with increasing dimension
- Bootstrapping and sample splitting for high-dimensional, assumption-lean inference
- Bootstrapping Lasso estimators
- Bounds for the asymptotic normality of the maximum likelihood estimator using the Delta method
- Bounds for the normal approximation of the maximum likelihood estimator
- CAN ONE ESTIMATE THE UNCONDITIONAL DISTRIBUTION OF POST-MODEL-SELECTION ESTIMATORS?
- Central limit theorems and bootstrap in high dimensions
- Comparison and anti-concentration bounds for maxima of Gaussian random vectors
- Confidence Intervals and Hypothesis Testing for High-Dimensional Regression
- Confidence intervals for low dimensional parameters in high dimensional linear models
- Confidence sets in sparse regression
- Controlling the false discovery rate via knockoffs
- Discussion: ``A significance test for the lasso
- Distribution-free predictive inference for regression
- Estimation and accuracy after model selection
- Exact post-selection inference, with application to the Lasso
- Frequentist Model Average Estimators
- Gaussian approximations and multiplier bootstrap for maxima of sums of high-dimensional random vectors
- Goodness-of-Fit Tests for High Dimensional Linear Models
- Group Bound: Confidence Intervals for Groups of Variables in Sparse High Dimensional Regression Without Assumptions on the Design
- High-dimensional inference in misspecified linear models
- High-dimensional inference: confidence intervals, \(p\)-values and R-software \texttt{hdi}
- High-dimensional simultaneous inference with the bootstrap
- High-dimensional variable selection
- Honest confidence regions for nonparametric regression
- scientific article; zbMATH DE number 6378123 (Why is no real title available?)
- scientific article; zbMATH DE number 46873 (Why is no real title available?)
- scientific article; zbMATH DE number 1975229 (Why is no real title available?)
- scientific article; zbMATH DE number 3446212 (Why is no real title available?)
- Hybrid and Size-Corrected Subsampling Methods
- Linear Model Selection by Cross-Validation
- Lower bounds for the rate of convergence in the central limit theorem in Banach spaces
- Models as approximations. I. Consequences illustrated with linear regression
- Multivariate normal approximation of the maximum likelihood estimator via the delta method
- Normal approximation for nonlinear statistics using a concentration inequality approach
- On asymptotically optimal confidence regions and tests for high-dimensional models
- Optimal-order bounds on the rate of convergence to normality in the multivariate delta method
- Panning for Gold: ‘Model-X’ Knockoffs for High Dimensional Controlled Variable Selection
- Probability for Statisticians
- Quantifying uncertainty in random forests via confidence intervals and hypothesis tests
- Random design analysis of ridge regression
- Rates of convergence of the adaptive LASSO estimators to the oracle distribution and higher order refinements by the bootstrap
- Selective inference with a randomized response
- Stability selection. With discussion and authors' reply
- Statistical significance in high-dimensional linear models
- Stein's method for nonlinear statistics: a brief survey and recent progress
- Uniform asymptotic inference and the bootstrap after model selection
- Uniform post-selection inference for least absolute deviation regression and other Z-estimation problems
- Uniformly valid confidence intervals post-model-selection
- Valid confidence intervals for post-model-selection predictors
- Valid post-selection inference
- Variable Selection with Error Control: Another Look at Stability Selection
Cited in
(43)- Testing conditional independence in supervised learning algorithms
- Uniform asymptotic inference and the bootstrap after model selection
- High-dimensional simultaneous inference with the bootstrap
- Multicarving for high-dimensional post-selection inference
- Exact minimax risk for linear least squares, and the lower tail of sample covariance matrices
- Averaging p-values under exchangeability
- Post-model-selection inference in linear regression models: an integrated review
- Comment: Statistical inference from a predictive perspective
- Rejoinder: On nearly assumption-free tests of nominal confidence interval coverage for causal parameters estimated by machine learning
- Bootstrapping and sample splitting for high-dimensional, assumption-lean inference
- Analyzing large datasets with bootstrap penalization
- \(p\)-values for high-dimensional regression
- Comments on: ``High-dimensional simultaneous inference with the bootstrap
- Comments on: ``High-dimensional simultaneous inference with the bootstrap
- Comments on: ``High-dimensional simultaneous inference with the bootstrap
- Mixed-effect models with trees
- Bootstrapping some GLM and survival regression variable selection estimators
- UNIFORM-IN-SUBMODEL BOUNDS FOR LINEAR REGRESSION IN A MODEL-FREE FRAMEWORK
- Synthetic learner: model-free inference on treatments over time
- Dimension-agnostic inference using cross U-statistics
- Post-selection inference via algorithmic stability
- A stable and adaptive polygenic signal detection method based on repeated sample splitting
- A Two-Sample Conditional Distribution Test Using Conformal Prediction and Weighted Rank Sum
- Feature importance: a closer look at Shapley values and LOCO
- Randomized empirical likelihood test for ultra-high dimensional means under general covariances
- Robust model averaging prediction of longitudinal response with ultrahigh-dimensional covariates
- Self-normalized Cramér type moderate deviation theorem for Gaussian approximation
- Dimension-agnostic change point detection
- A unified framework for covariate adjustment under stratified randomisation
- Data Fission: Splitting a Single Data Point
- Debiased Lasso after sample splitting for estimation and inference in high-dimensional generalized linear models
- Exact selective inference with randomization
- Improving power by conditioning on less in post-selection inference for changepoints
- Post-selection inference in regression models for group testing data
- One component partial least squares, high dimensional regression, data splitting, and the multitude of models
- The projected covariance measure for assumption-lean variable significance testing
- Data Thinning for Poisson Factor Models and its Applications
- Robust universal inference for misspecified models
- The limits of assumption-free tests for algorithm performance
- Spectral clustering with likelihood refinement for high-dimensional latent class recovery
- Group regression with tailored penalties for high-dimensional data with or without predefined groups
- Double cross-fit doubly robust estimators: beyond series regression
- Two-directional simultaneous inference for high-dimensional models
This page was built for publication: Bootstrapping and sample splitting for high-dimensional, assumption-lean inference
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2284380)
