P-values for high-dimensional regression

Abstract

Assigning significance in high-dimensional regression is challenging. Most computationally efficient selection algorithms cannot guard against inclusion of noise variables. Asymptotically valid p-values are not available. An exception is a recent proposal by Wasserman and Roeder (2008) which splits the data into two parts. The number of variables is then reduced to a manageable size using the first split, while classical variable selection techniques can be applied to the remaining variables, using the data from the second split. This yields asymptotic error control under minimal conditions. It involves, however, a one-time random split of the data. Results are sensitive to this arbitrary choice: it amounts to a `p-value lottery' and makes it difficult to reproduce results. Here, we show that inference across multiple random splits can be aggregated, while keeping asymptotic control over the inclusion of noise variables. We show that the resulting p-values can be used for control of both family-wise error (FWER) and false discovery rate (FDR). In addition, the proposed aggregation is shown to improve power while reducing the number of falsely selected variables substantially.

25 pages, 4 figures

References in corpus (1)

• Stability Selection

Cited by in corpus (14)

• A significance test for the lasso
• SLOPE - Adaptive variable selection via convex optimization
• Statistical significance in high-dimensional linear models
• Inference and Uncertainty Quantification for Noisy Matrix Completion
• High-dimensional inference in misspecified linear models
• Goodness of fit tests for high-dimensional linear models
• Wavelet-domain regression and predictive inference in psychiatric neuroimaging
• Significance tests of feature relevance for a black-box learner
• Type I error rate control for testing many hypotheses: a survey with proofs
• A Two-Sample Conditional Distribution Test Using Conformal Prediction and Weighted Rank Sum
• Effect modification in anchored indirect treatment comparisons: Comments on "Matching-adjusted indirect comparisons: Application to time-to-event data"
• The Smooth-Lasso and other ${\ell }_1+{\ell }_2$-penalized methods
• False Discovery Proportion control for aggregated Knockoffs
• Exact model comparisons in the plausibility framework