Abstrakt
Double machine learning (DML) is becoming an increasingly popular tool for automated 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.
Originalsprog | Engelsk |
---|---|
Titel | Proceedings of the Eighty-second Annual Meeting of the Academy of Management |
Redaktører | Sonia Taneja |
Antal sider | 1 |
Udgivelsessted | Briarcliff Manor, NY |
Forlag | Academy of Management |
Publikationsdato | 2022 |
Sider | 2199 |
DOI | |
Status | Udgivet - 2022 |
Begivenhed | The Academy of Management Annual Meeting 2022: Creating a Better World Together - Seattle, USA Varighed: 5 aug. 2022 → 9 aug. 2022 Konferencens nummer: 82 https://2022.aom.org/ |
Konference
Konference | The Academy of Management Annual Meeting 2022 |
---|---|
Nummer | 82 |
Land/Område | USA |
By | Seattle |
Periode | 05/08/2022 → 09/08/2022 |
Internetadresse |
Navn | Academy of Management Proceedings |
---|---|
ISSN | 0065-0668 |