mlsanityData to 7 Oct 2026

Meta-Learners for Estimation of Causal Effects: Finite Sample Cross-Fit Performance

arXiv 2201.12692 ↗First posted 30 Jan 2022econ.EM stat.MLcited by 4

Not on the map. Why?

Abstract (arXiv)

Estimation of causal effects using machine learning methods has become an active research field in econometrics. In this paper, we study the finite sample performance of meta-learners for estimation of heterogeneous treatment effects under the usage of sample-splitting and cross-fitting to reduce the overfitting bias. In both synthetic and semi-synthetic simulations we find that the performance of the meta-learners in finite samples greatly depends on the estimation procedure. The results imply that sample-splitting and cross-fitting are beneficial in large samples for bias reduction and efficiency of the meta-learners, respectively, whereas full-sample estimation is preferable in small samples. Furthermore, we derive practical recommendations for application of specific meta-learners in empirical studies depending on particular data characteristics such as treatment shares and sample size.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 39 of its 146 references that are in the map. First posted on arXiv 30 Jan 2022.

Builds on

39 of 146 references.

References · 39

  1. 2015Estimation and Inference of Heterogeneous Treatment Effects using Random Forestscited by 597
  2. 2017Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learningcited by 475
  3. 2017Quasi-Oracle Estimation of Heterogeneous Treatment Effectscited by 328
  4. 2016Learning Representations for Counterfactual Inferencecited by 327
  5. 2016Generalized Random Forestscited by 309
  6. 2019Adapting Neural Networks for the Estimation of Treatment Effectscited by 285
  7. 2015Recursive Partitioning for Heterogeneous Causal Effectscited by 254
  8. 2015ranger: A Fast Implementation of Random Forests for High Dimensional Data in C++ and Rcited by 171
  9. 2015A Random Forest Guided Tourcited by 168
  10. 2021Nonparametric Estimation of Heterogeneous Treatment Effects: From Theory to Learning Algorithmscited by 144
  11. 2014Consistency of random forestscited by 84
  12. 2021Discussion on "Bayesian Regression Tree Models for Causal Inference: Regularization, Confounding, and Heterogeneous Effects" by Hahn, Murray and Carvalhocited by 77
27 more
  1. 2019Machine Learning Methods Economists Should Know Aboutcited by 63
  2. 2015Nonparametric methods for doubly robust estimation of continuous treatment effectscited by 62
  3. 2018Perfect Match: A Simple Method for Learning Representations For Counterfactual Inference With Neural Networkscited by 55
  4. 2019Learning When-to-Treat Policiescited by 49
  5. 2017Some methods for heterogeneous treatment effect estimation in high-dimensionscited by 48
  6. 2013Confidence Intervals for Random Forests: The Jackknife and the Infinitesimal Jackknifecited by 43
  7. 2016The State of Applied Econometrics - Causality and Policy Evaluationcited by 41
  8. 2017Bayesian regression tree models for causal inference: regularization, confounding, and heterogeneous effectscited by 41
  9. 2020A unified survey of treatment effect heterogeneity modeling and uplift modelingcited by 36
  10. 2020How Important is the Train-Validation Split in Meta-Learning?cited by 30
  11. 2013Program Evaluation and Causal Inference with High-Dimensional Datacited by 29
  12. 2018A comparison of methods for model selection when estimating individual treatment effectscited by 28
  13. 2017Estimating Individual Treatment Effect in Observational Data Using Random Forest Methodscited by 21
  14. 2019Response Transformation and Profit Decomposition for Revenue Uplift Modelingcited by 15
  15. 2019Assessing Treatment Effect Variation in Observational Studies: Results from a Data Challengecited by 12
  16. 2021A Representation Learning Perspective on the Importance of Train-Validation Splitting in Meta-Learningcited by 12
  17. 2019Machine learning in policy evaluation: new tools for causal inferencecited by 10
  18. 2020Machine learning for causal inference: on the use of cross-fit estimatorscited by 10
  19. 2017Challenges in Obtaining Valid Causal Effect Estimates with Machine Learning Algorithmscited by 8
  20. 2020Minimax Optimal Nonparametric Estimation of Heterogeneous Treatment Effectscited by 8
  21. 2016Doubly Robust Uniform Confidence Band for the Conditional Average Treatment Effect Functioncited by 6
  22. 2018Modified Causal Forests for Estimating Heterogeneous Causal Effectscited by 5
  23. 2019Priority to unemployed immigrants? A causal machine learning evaluation of training in Belgiumcited by 5
  24. 2020Cross-Fitting and Averaging for Machine Learning Estimation of Heterogeneous Treatment Effectscited by 5
  25. 2014Asymptotic Theory for Random Forestscited by 4
  26. 2019Nonparametric Heterogeneous Treatment Effect Estimation in Repeated Cross Sectional Designscited by 4
  27. 2017Inference on a New Class of Sample Average Treatment Effectscited by 1

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 0; 2024: 0; 2025: 3.

4 papers.

  1. 2022Comparison of meta-learners for estimating multi-valued treatment heterogeneous effectscited by 10
  2. 2025Multiple Treatments Causal Effects Estimation with Task Embeddings and Balanced Representation Learningcited by 1
  3. 2025Scalable Vision-Guided Crop Yield Estimationcited by 1
  4. 2025Assessment of the conditional exchangeability assumption in causal machine learning models: a simulation studycited by 0