A robustness test for estimating total effects with covariate adjustment

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research

Standard

A robustness test for estimating total effects with covariate adjustment. / Su, Zehao; Henckel, Leonard.

Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence. 1. ed. PMLR, 2022. p. 1886-1895 (Proceedings of Machine Learning Research, Vol. 180).

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research

Harvard

Su, Z & Henckel, L 2022, A robustness test for estimating total effects with covariate adjustment. in Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence. 1 edn, PMLR, Proceedings of Machine Learning Research, vol. 180, pp. 1886-1895, 38th Conference on Uncertainty in Artificial Intelligence (UAI 2022), Eindhoven, Netherlands, 01/08/2022. <https://proceedings.mlr.press/v180/su22a.html>

APA

Su, Z., & Henckel, L. (2022). A robustness test for estimating total effects with covariate adjustment. In Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence (1 ed., pp. 1886-1895). PMLR. Proceedings of Machine Learning Research Vol. 180 https://proceedings.mlr.press/v180/su22a.html

Vancouver

Su Z, Henckel L. A robustness test for estimating total effects with covariate adjustment. In Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence. 1 ed. PMLR. 2022. p. 1886-1895. (Proceedings of Machine Learning Research, Vol. 180).

Author

Su, Zehao ; Henckel, Leonard. / A robustness test for estimating total effects with covariate adjustment. Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence. 1. ed. PMLR, 2022. pp. 1886-1895 (Proceedings of Machine Learning Research, Vol. 180).

Bibtex

@inproceedings{212241fd86644124ba6cdbe0aa57daab,
title = "A robustness test for estimating total effects with covariate adjustment",
abstract = "Suppose we want to estimate a total effect with covariate adjustment in a linear structural equation model. We have a causal graph to decide what covariates to adjust for, but are uncertain about the graph. Here, we propose a testing procedure, that exploits the fact that there are multiple valid adjustment sets for the target total effect in the causal graph, to perform a robustness check on the graph. If the test rejects, it is a strong indication that we should not rely on the graph. We discuss what mistakes in the graph our testing procedure can detect and which ones it cannot and develop two strategies on how to select a list of valid adjustment sets for the procedure. We also connect our result to the related econometrics literature on coefficient stability tests",
author = "Zehao Su and Leonard Henckel",
year = "2022",
language = "English",
series = "Proceedings of Machine Learning Research",
pages = "1886--1895",
booktitle = "Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence",
publisher = "PMLR",
edition = "1",
note = "38th Conference on Uncertainty in Artificial Intelligence (UAI 2022) ; Conference date: 01-08-2022 Through 05-08-2022",

}

RIS

TY - GEN

T1 - A robustness test for estimating total effects with covariate adjustment

AU - Su, Zehao

AU - Henckel, Leonard

PY - 2022

Y1 - 2022

N2 - Suppose we want to estimate a total effect with covariate adjustment in a linear structural equation model. We have a causal graph to decide what covariates to adjust for, but are uncertain about the graph. Here, we propose a testing procedure, that exploits the fact that there are multiple valid adjustment sets for the target total effect in the causal graph, to perform a robustness check on the graph. If the test rejects, it is a strong indication that we should not rely on the graph. We discuss what mistakes in the graph our testing procedure can detect and which ones it cannot and develop two strategies on how to select a list of valid adjustment sets for the procedure. We also connect our result to the related econometrics literature on coefficient stability tests

AB - Suppose we want to estimate a total effect with covariate adjustment in a linear structural equation model. We have a causal graph to decide what covariates to adjust for, but are uncertain about the graph. Here, we propose a testing procedure, that exploits the fact that there are multiple valid adjustment sets for the target total effect in the causal graph, to perform a robustness check on the graph. If the test rejects, it is a strong indication that we should not rely on the graph. We discuss what mistakes in the graph our testing procedure can detect and which ones it cannot and develop two strategies on how to select a list of valid adjustment sets for the procedure. We also connect our result to the related econometrics literature on coefficient stability tests

M3 - Article in proceedings

T3 - Proceedings of Machine Learning Research

SP - 1886

EP - 1895

BT - Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence

PB - PMLR

T2 - 38th Conference on Uncertainty in Artificial Intelligence (UAI 2022)

Y2 - 1 August 2022 through 5 August 2022

ER -

ID: 334858619