EconBase
← All papers

Robust Orthogonal Machine Learning of Treatment Effects

Yiyan Huang, Cheuk Hang Leung, Qi Wu, Xing Yan

arXiv 22 Mar 2021 · Statistics — Machine Learning

arXiv:2103.11869 · PDF · DOI · OpenAlex · Extracted main text

Abstract

Causal learning is the key to obtaining stable predictions and answering what if problems in decision-makings. In causal learning, it is central to seek methods to estimate the average treatment effect (ATE) from observational data. The Double/Debiased Machine Learning (DML) is one of the prevalent methods to estimate ATE. However, the DML estimators can suffer from an error-compounding issue and even give extreme estimates when the propensity scores are close to 0 or 1. Previous studies have overcome this issue through some empirical tricks such as propensity score trimming, yet none of the existing works solves it from a theoretical standpoint. In this paper, we propose a Robust Causal Learning (RCL) method to offset the deficiencies of DML estimators. Theoretically, the RCL estimators i) satisfy the (higher-order) orthogonal condition and are as consistent and doubly robust as the DML estimators, and ii) get rid of the error-compounding issue. Empirically, the comprehensive experiments show that: i) the RCL estimators give more stable estimations of the causal parameters than DML; ii) the RCL estimators outperform traditional estimators and their variants when applying different machine learning models on both simulation and benchmark datasets, and a mimic consumer credit dataset generated by WGAN.

Citation extraction

48
references
87
in-text mentions
48
distinct cited
0
self-citations
19,553
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
1V. Chernozhukov, D. Chetverikov, M. Demirer, E. Duflo, C. Hansen, W.… (2018) Double/debiased machine learning for treatment and structural parameters1.000134100%
2L. Mackey, V. Syrgkanis, and I. Zadik, “Orthogonal machine learning:… (2018) Orthogonal machine learning: Power and limitations1.000105100%
3J. L. Hill, “Bayesian nonparametric modeling for causal inference,”… (2011) Bayesian nonparametric modeling for causal inference0.81142100%
4U. Shalit, F. D. Johansson, and D. Sontag, “Estimating individual tr… (2017) Estimating individual treatment effect: generalization bounds and algorithms0.81142100%
5C. Shi, D. Blei, and V. Veitch, “Adapting neural networks for the es… (2019) Adapting neural networks for the estimation of treatment effects0.81142100%
6J. Yoon, J. Jordon, and M. Van Der Schaar, “Ganite: Estimation of in… (2018) Ganite: Estimation of individualized treatment effects using generative adversarial nets0.73732100%
7A. Linden, S. D. Uysal, A. Ryan, and J. L. Adams, “Estimating causal… (2016) Estimating causal effects for multivalued treatments: a comparison of approaches0.64422100%
8C. Louizos, U. Shalit, J. M. Mooij, D. Sontag, R. Zemel, and M. Well… (2017) Causal effect inference with deep latent-variable models0.64422100%
9J. Robins, L. Li, E. Tchetgen, A. van der Vaart et al., “Higher orde… (2008) Higher order influence functions and minimax estimation of nonlinear functionals0.64422100%
10S. Athey, G. W. Imbens, J. Metzger, and E. Munro, “Using wasserstein… (2021) Using wasserstein generative adversarial networks for the design of monte carlo simulations0.58531100%

Showing the top 10 of 48 scored citations.