Double Machine Learning and Automated Confounder Selection -- A Cautionary Tale
Double machine learning (DML) is becoming an increasingly popular tool for automatic model selection in high-dimensional settings. These approaches rely on the assumption of conditional independence, which may not hold in big-data settings where the covariate space is large. This paper shows that DM...
Main Authors: | , , |
---|---|
Format: | Text |
Language: | unknown |
Published: |
2021
|
Subjects: | |
Online Access: | http://arxiv.org/abs/2108.11294 |
Summary: | Double machine learning (DML) is becoming an increasingly popular tool for automatic model selection in high-dimensional settings. These approaches rely on the assumption of conditional independence, which may not hold in big-data settings where the covariate space is large. This paper shows that DML is very sensitive to the inclusion of even a few "bad controls" in the covariate space. The resulting bias varies with the nature of the causal model, which raises concerns about the feasibility of selecting control variables in a data-driven way. |
---|