EconBase
← All papers

Automatic Debiased Machine Learning via Riesz Regression

Victor Chernozhukov, Whitney K. Newey, Victor Quintas-Martinez, Vasilis Syrgkanis

arXiv 30 Apr 2021 · Mathematics — Statistics Theory · 9 citations (OpenAlex)

arXiv:2104.14737 · PDF · DOI · OpenAlex · Extracted main text

Abstract

A variety of interesting parameters may depend on high dimensional regressions. Machine learning can be used to estimate such parameters. However estimators based on machine learners can be severely biased by regularization and/or model selection. Debiased machine learning uses Neyman orthogonal estimating equations to reduce such biases. Debiased machine learning generally requires estimation of unknown Riesz representers. A primary innovation of this paper is to provide Riesz regression estimators of Riesz representers that depend on the parameter of interest, rather than explicit formulae, and that can employ any machine learner, including neural nets and random forests. End-to-end algorithms emerge where the researcher chooses the parameter of interest and the machine learner and the debiasing follows automatically. Another innovation here is debiased machine learners of parameters depending on generalized regressions, including high-dimensional generalized linear models. An empirical example of automatic debiased machine learning using neural nets is given. We find in Monte Carlo examples that automatic debiasing sometimes performs better than debiasing via inverse propensity scores and never worse. Finite sample mean square error bounds for Riesz regression estimators and asymptotic theory are also given.

Citation extraction

38
references
75
in-text mentions
38
distinct cited
5
self-citations
17,094
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
1barticle[author] Farrell, Max H.M. H., Liang, TengyuanT. Misra, Sanj… (2021) a)1.00094100%
2barticle[author] Chernozhukov, VictorV., Escanciano, Juan CarlosJ. C… (2022) b)1.00064100%
3binproceedings[author] Chernozhukov, VictorV., Newey, Whitney K.W. K… (2022) a) self0.92844100%
4barticle[author] Ichimura, HidehikoH. Newey, Whitney KW. K (2022) ) self0.87452100%
5barticle[author] Foster, Dylan JD. J. Syrgkanis, VasilisV (2019) ) self0.84333100%
6barticle[author] Newey, Whitney K.W. K (1994) )0.84333100%
7binproceedings[author] Yarotsky, DmitryD (2018) )0.81142100%
8barticle[author] Hirshberg, David A.D. A. Wager, StefanS (2021) )0.73732100%
9bmisc[author] Chernozhukov, VictorV., Newey, WhitneyW. Singh, RahulR (2019) )0.64422100%
10barticle[author] Chernozhukov, VictorV., Newey, Whitney K.W. K. Sing… (2022) )0.64422100%

Showing the top 10 of 38 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
1Welfare Analysis in Dynamic Models0.79485
2Multiply-Robust Causal Change Attribution0.64422
3Sensitivity Analysis for Treatment Effects in Difference-in-Differences Models using Riesz Representation0.64422
4Sharp Structure-Agnostic Lower Bounds for General Linear Functional Estimation0.64422
5A Locally Robust Semiparametric Approach to Examiner IV Designs0.51122
6Two-Step Estimation of a Strategic Network Formation Model with Clustering0.40511
7Deep Learning for Individual Heterogeneity0.40511
8Difference-in-Differences for Continuous Treatments and Instruments with Stayers0.40511
9Automatic Locally Robust GMM with Machine-Learning-Generated Regressors0.40511
10Causal Inference in High-Dimensional Generalized Linear Models with Binary Outcomes0.40511