Omitted Variable Bias of Lasso-Based Inference Methods Under Limited Variability: A Finite Sample Analysis

Wüthrich, Kaspar; Zhu, Ying

doi:10.2139/ssrn.3379123

Cited by 4 publications

(2 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In a Monte Carlo study, Knaus, Lechner and Strittmatter (2018) compare ML-based estimates of individual average treatment effects, focusing on effect heterogeneity. We discuss a related paper by Wuthrich and Zhu (2019) below.…”

Section: Introductionmentioning

confidence: 99%

Machine Labor

Angrist¹,

Frandsen²

2019

View full text Add to dashboard Cite

Walters, and seminar participants at Columbia and MIT for helpful discussions and comments. They are not to blame for any of our mistakes or conclusions. This paper is dedicated to the memory of Alan Krueger. We would have liked to have his comments. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peerreviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.

show abstract

Section: Introductionmentioning

confidence: 99%

Machine Labor

Angrist¹,

Frandsen²

2019

View full text Add to dashboard Cite

show abstract

“…But in most realistic settings, some confounders remain unadjusted for, raising the prospect of such biases being merely reduced, but not entirely eliminated. For example, there may be either confounders that are unfeasible or impractical to observe or record 31 , or (measured) confounders excluded by routine variable selection techniques 40 . It is therefore crucial that investigators can practically assess the relative stability of their conclusions to dormant confounding that remains unadjusted for.…”

Section: Introductionmentioning

confidence: 99%

Sensitivity Analysis for Unmeasured Confounding via Effect Extrapolation

Loh,

Vansteelandt

2021

Preprint

View full text Add to dashboard Cite

Inferring the causal effect of a non-randomly assigned exposure on an outcome requires adjusting for common causes of the exposure and outcome to avoid biased conclusions. Notwithstanding the efforts investigators routinely make to measure and adjust for such common causes (or confounders), some confounders typically remain unmeasured, raising the prospect of biased inference in observational studies. Therefore, it is crucial that investigators can practically assess their substantive conclusions' relative (in)sensitivity to potential unmeasured confounding. In this article, we propose a sensitivity analysis strategy that is informed by the stability of the exposure effect over different, wellchosen subsets of the measured confounders. The proposal entails first approximating the process for recording confounders to learn about how the effect is potentially affected by varying amounts of unmeasured confounding, then extrapolating to the effect had hypothetical unmeasured confounders been additionally adjusted for. A large set of measured confounders can thus be exploited to provide insight into the likely presence of unmeasured confounding bias, albeit under an assumption about how data on the confounders are recorded. The proposal's ability to reveal the true effect and ensure valid inference after extrapolation is empirically compared with existing methods using simulation studies. We demonstrate the procedure using two different publicly available datasets commonly used for causal inference.

show abstract

Double machine learning and automated confounder selection: A cautionary tale

Hünermund

Louw

Caspi³

2023

Journal of Causal Inference

View full text Add to dashboard Cite

Double machine learning (DML) has become an increasingly popular tool for automated variable selection in high-dimensional settings. Even though the ability to deal with a large number of potential covariates can render selection-on-observables assumptions more plausible, there is at the same time a growing risk that endogenous variables are included, which would lead to the violation of conditional independence. This article demonstrates that DML is very sensitive to the inclusion of only a few “bad controls” in the covariate space. The resulting bias varies with the nature of the theoretical causal model, which raises concerns about the feasibility of selecting control variables in a data-driven way.

show abstract

Omitted Variable Bias of Lasso-Based Inference Methods Under Limited Variability: A Finite Sample Analysis

Cited by 4 publications

References 49 publications

Machine Labor

Machine Labor

Sensitivity Analysis for Unmeasured Confounding via Effect Extrapolation

Double machine learning and automated confounder selection: A cautionary tale

Contact Info

Product

Resources

About