Advanced search
2 files | 1.08 MB Add to list

Doubly robust tests of exposure effects under high-dimensional confounding

(2020) BIOMETRICS. 76(4). p.1190-1200
Author
Organization
Abstract
After variable selection, standard inferential procedures for regression parameters may not be uniformly valid; there is no finite-sample size at which a standard test is guaranteed to approximately attain its nominal size. This problem is exacerbated in high-dimensional settings, where variable selection becomes unavoidable. This has prompted a flurry of activity in developing uniformly valid hypothesis tests for a low-dimensional regression parameter (eg, the causal effect of an exposure A on an outcome Y) in high-dimensional models. So far there has been limited focus on model misspecification, although this is inevitable in high-dimensional settings. We propose tests of the null that are uniformly valid under sparsity conditions weaker than those typically invoked in the literature, assuming working models for the exposure and outcome are both correctly specified. When one of the models is misspecified, by amending the procedure for estimating the nuisance parameters, our tests continue to be valid; hence, they are doubly robust. Our proposals are straightforward to implement using existing software for penalized maximum likelihood estimation and do not require sample splitting. We illustrate them in simulations and an analysis of data obtained from the Ghent University intensive care unit.
Keywords
General Biochemistry, Genetics and Molecular Biology, Statistics and Probability, General Immunology and Microbiology, Applied Mathematics, General Agricultural and Biological Sciences, General Medicine, causal inference, doubly robust estimation, high-dimensional inference, post-selection inference, CONFIDENCE-REGIONS, INFERENCE, SELECTION

Downloads

  • Biometrics HDDR SM revision1.pdf
    • supplementary material
    • |
    • open access
    • |
    • PDF
    • |
    • 613.82 KB
  • HDDR manuscript.pdf
    • full text (Accepted manuscript)
    • |
    • open access
    • |
    • PDF
    • |
    • 463.71 KB

Citation

Please use this url to cite or link to this publication:

MLA
Dukes, Oliver, et al. “Doubly Robust Tests of Exposure Effects under High-Dimensional Confounding.” BIOMETRICS, vol. 76, no. 4, 2020, pp. 1190–200, doi:10.1111/biom.13231.
APA
Dukes, O., Avagyan, V., & Vansteelandt, S. (2020). Doubly robust tests of exposure effects under high-dimensional confounding. BIOMETRICS, 76(4), 1190–1200. https://doi.org/10.1111/biom.13231
Chicago author-date
Dukes, Oliver, Vahe Avagyan, and Stijn Vansteelandt. 2020. “Doubly Robust Tests of Exposure Effects under High-Dimensional Confounding.” BIOMETRICS 76 (4): 1190–1200. https://doi.org/10.1111/biom.13231.
Chicago author-date (all authors)
Dukes, Oliver, Vahe Avagyan, and Stijn Vansteelandt. 2020. “Doubly Robust Tests of Exposure Effects under High-Dimensional Confounding.” BIOMETRICS 76 (4): 1190–1200. doi:10.1111/biom.13231.
Vancouver
1.
Dukes O, Avagyan V, Vansteelandt S. Doubly robust tests of exposure effects under high-dimensional confounding. BIOMETRICS. 2020;76(4):1190–200.
IEEE
[1]
O. Dukes, V. Avagyan, and S. Vansteelandt, “Doubly robust tests of exposure effects under high-dimensional confounding,” BIOMETRICS, vol. 76, no. 4, pp. 1190–1200, 2020.
@article{8664944,
  abstract     = {{After variable selection, standard inferential procedures for regression parameters may not be uniformly valid; there is no finite-sample size at which a standard test is guaranteed to approximately attain its nominal size. This problem is exacerbated in high-dimensional settings, where variable selection becomes unavoidable. This has prompted a flurry of activity in developing uniformly valid hypothesis tests for a low-dimensional regression parameter (eg, the causal effect of an exposure A on an outcome Y) in high-dimensional models. So far there has been limited focus on model misspecification, although this is inevitable in high-dimensional settings. We propose tests of the null that are uniformly valid under sparsity conditions weaker than those typically invoked in the literature, assuming working models for the exposure and outcome are both correctly specified. When one of the models is misspecified, by amending the procedure for estimating the nuisance parameters, our tests continue to be valid; hence, they are doubly robust. Our proposals are straightforward to implement using existing software for penalized maximum likelihood estimation and do not require sample splitting. We illustrate them in simulations and an analysis of data obtained from the Ghent University intensive care unit.}},
  author       = {{Dukes, Oliver and Avagyan, Vahe and Vansteelandt, Stijn}},
  issn         = {{0006-341X}},
  journal      = {{BIOMETRICS}},
  keywords     = {{General Biochemistry,Genetics and Molecular Biology,Statistics and Probability,General Immunology and Microbiology,Applied Mathematics,General Agricultural and Biological Sciences,General Medicine,causal inference,doubly robust estimation,high-dimensional inference,post-selection inference,CONFIDENCE-REGIONS,INFERENCE,SELECTION}},
  language     = {{eng}},
  number       = {{4}},
  pages        = {{1190--1200}},
  title        = {{Doubly robust tests of exposure effects under high-dimensional confounding}},
  url          = {{http://dx.doi.org/10.1111/biom.13231}},
  volume       = {{76}},
  year         = {{2020}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: