EconBase
← All papers

On the Asymptotic Properties of Debiased Machine Learning Estimators

Amilcar Velez

arXiv 4 Nov 2024 · Econometrics

arXiv:2411.01864 · PDF · DOI · OpenAlex · Extracted main text

Abstract

This paper studies the properties of debiased machine learning (DML) estimators under a novel asymptotic framework, offering insights for improving the performance of these estimators in applications. DML is an estimation method suited to economic models where the parameter of interest depends on unknown nuisance functions that must be estimated. It requires weaker conditions than previous methods while still ensuring standard asymptotic properties. Existing theoretical results do not distinguish between two alternative versions of DML estimators, DML1 and DML2. Under a new asymptotic framework, this paper demonstrates that DML2 asymptotically dominates DML1 in terms of bias and mean squared error, formalizing a previous conjecture based on simulation results regarding their relative performance. Additionally, this paper provides guidance for improving the performance of DML2 in applications.

Citation extraction

54
references
146
in-text mentions
54
distinct cited
0
self-citations
51,797
main-text words

appendix boundary found by none_found · 100% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1Chernozhukov, V., D. Chetverikov, M. Demirer, E. Duflo, C. Hansen, W… (2018) Double/debiased machine learning for treatment and structural parameters1.000126100%
2Sant’Anna, P. H. and J. Zhao (2020) Doubly robust difference-in-differences estimators1.00083100%
3Linton, O (1995) Second order approximation in the partially linear regression model1.00064100%
4Newey, W. K. and R. J. Smith (2004) Higher order properties of GMM and generalized empirical likelihood estimators1.00063100%
5Andrews, D. W (1994) Asymptotics for semiparametric econometric models via stochastic equicontinuity1.00053100%
6Hirano, K., G. W. Imbens, and G. Ridder (2003) Efficient estimation of average treatment effects using the estimated propensity score1.00053100%
7Chernozhukov, V., J. C. Escanciano, H. Ichimura, W. K. Newey, and J.… (2022) a): Locally robust semiparametric estimation0.92843100%
8Newey, W. K (1994) The asymptotic variance of semiparametric estimators0.92843100%
9Ahrens, A., C. B. Hansen, M. E. Schaffer, and T. Wiemann (2024) a): ddml: Double/debiased machine learning in Stata0.84333100%
10Ahrens, A., C. B. Hansen, M. E. Schaffer, and T. Wiemann (2024) b): Model averaging and double machine learning0.84333100%

Showing the top 10 of 54 scored citations.

Cited by, within the corpus

arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.

Citing paperIntensityMentionsSections
1An Introduction to Double/Debiased Machine Learning0.73732
2Inferring Treatment Effects in Large Panels by Uncovering Latent Similarities0.64422
3Deep Learning for Individual Heterogeneity0.40511
4Flexible Covariate Adjustments in Regression Discontinuity DesignsFirst version: July 16, 2021. This version: . We thank Sebastian Calonico, Michal Kolesár, Thomas Lemieux, Jonathan Roth, Vira Semenova, Stefan Wager, Daniel Wilhelm, Andrei Zeleneev, and numerous conference and seminar participants for helpful comments and suggestions. We thank Tobias Grobölting and Merve Ögretmek for excellent research assistance. The authors gratefully acknowledge financial support by the European Research Council (ERC) through grant SH1-77202. The second author also gratefully acknowledges support from the European Research Council ERC through grant SH-1852332. Author contact information: Claudia Noack, Department of Economics, University of Bonn0.40511
5Lee Bounds with a Continuous Treatment in Sample Selection0.40511
6Leave No One Undermined: Policy Targeting with Regret Aversion0.40511
7Training and Testing with Multiple Splits: A Central Limit Theorem for Split-Sample Estimators0.40511
8Identification and Inference for Algorithmic Frontiers with Selective Labels0.40511