ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 1909.10828
6
10

Double-estimation-friendly inference for high-dimensional misspecified models

24 September 2019
Rajen Dinesh Shah
Peter Buhlmann
ArXivPDFHTML
Abstract

All models may be wrong -- but that is not necessarily a problem for inference. Consider the standard ttt-test for the significance of a variable XXX for predicting response YYY whilst controlling for ppp other covariates ZZZ in a random design linear model. This yields correct asymptotic type~I error control for the null hypothesis that XXX is conditionally independent of YYY given ZZZ under an \emph{arbitrary} regression model of YYY on (X,Z)(X, Z)(X,Z), provided that a linear regression model for XXX on ZZZ holds. An analogous robustness to misspecification, which we term the "double-estimation-friendly" (DEF) property, also holds for Wald tests in generalised linear models, with some small modifications. In this expository paper we explore this phenomenon, and propose methodology for high-dimensional regression settings that respects the DEF property. We advocate specifying (sparse) generalised linear regression models for both YYY and the covariate of interest XXX; our framework gives valid inference for the conditional independence null if either of these hold. In the special case where both specifications are linear, our proposal amounts to a small modification of the popular debiased Lasso test. We also investigate constructing confidence intervals for the regression coefficient of XXX via inverting our tests; these have coverage guarantees even in partially linear models where the contribution of ZZZ to YYY can be arbitrary. Numerical experiments demonstrate the effectiveness of the methodology.

View on arXiv
Comments on this paper