High-dimensional inference in misspecified linear models

Abstract

We consider high-dimensional inference when the assumed linear model is misspecified. We describe some correct interpretations and corresponding sufficient assumptions for valid asymptotic inference of the model parameters, which still have a useful meaning when the model is misspecified. We largely focus on the de-sparsified Lasso procedure but we also indicate some implications for (multiple) sample splitting techniques. In view of available methods and software, our results contribute to robustness considerations with respect to model misspecification.

24 pages, 4 figures

References in corpus (5)

• High-dimensional variable selection
• P-values for high-dimensional regression
• Discussion: "A significance test for the lasso"
• Statistical Theory for High-Dimensional Models
• ${\chi }^2$-confidence sets in high-dimensional regression

Cited by in corpus (15)

• Linear Hypothesis Testing in Dense High-Dimensional Linear Models
• Goodness of fit tests for high-dimensional linear models
• DoubleML -- An Object-Oriented Implementation of Double Machine Learning in R
• A unifying approach for doubly-robust ${\ell }_1$ regularized estimation of causal contrasts
• Transfer Learning for High-dimensional Linear Regression: Prediction, Estimation, and Minimax Optimality
• Hierarchical inference for genome-wide association studies: a view on methodology with software
• DoubleML -- An Object-Oriented Implementation of Double Machine Learning in Python
• Floodgate: inference for model-free variable importance
• Semi-Supervised linear regression
• High Dimensional M-Estimation with Missing Outcomes: A Semi-Parametric Framework
• High-dimensional simultaneous inference with the bootstrap
• Incremental causal effects
• High-dimensional semi-supervised learning: in search for optimal inference of the mean
• Berry-Esseen Bounds for Projection Parameters and Partial Correlations with Increasing Dimension
• Inject Machine Learning into Significance Test for Misspecified Linear Models