mlsanityData to 14 Oct 2026

Stable Probability Weighting: Large-Sample and Finite-Sample Estimation and Inference Methods for Heterogeneous Causal Effects of Multivalued Treatments Under Limited Overlap

arXiv 2301.05703 ↗First posted 13 Jan 2023econ.EM math.ST stat.APcited by 0

Not on the map. Why?

Abstract (arXiv)

In this paper, I try to tame "Basu's elephants" (data with extreme selection on observables). I propose new practical large-sample and finite-sample methods for estimating and inferring heterogeneous causal effects (under unconfoundedness) in the empirically relevant context of limited overlap. I develop a general principle called "Stable Probability Weighting" (SPW) that can be used as an alternative to the widely used Inverse Probability Weighting (IPW) technique, which relies on strong overlap. I show that IPW (or its augmented version), when valid, is a special case of the more general SPW (or its doubly robust version), which adjusts for the extremeness of the conditional probabilities of the treatment states. The SPW principle can be implemented using several existing large-sample parametric, semiparametric, and nonparametric procedures for conditional moment models. In addition, I provide new finite-sample results that apply when unconfoundedness is plausible within fine strata. Since IPW estimation relies on the problematic reciprocal of the estimated propensity score, I develop a "Finite-Sample Stable Probability Weighting" (FPW) set-estimator that is unbiased in a sense. I also propose new finite-sample inference methods for testing a general class of weak null hypotheses. The associated computationally convenient methods, which can be used to construct valid confidence sets and to bound the finite-sample confidence distribution, are of independent interest. My large-sample and finite-sample frameworks extend to the setting of multivalued treatments.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 49 of its 172 references that are in the map. First posted on arXiv 13 Jan 2023.

Builds on

49 of 172 references.

References · 49

  1. 2015Estimation and Inference of Heterogeneous Treatment Effects using Random Forestscited by 600
  2. 2017Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learningcited by 478
  3. 2017Quasi-Oracle Estimation of Heterogeneous Treatment Effectscited by 332
  4. 2016Generalized Random Forestscited by 312
  5. 2015Recursive Partitioning for Heterogeneous Causal Effectscited by 255
  6. 2021Discussion on "Bayesian Regression Tree Models for Causal Inference: Regularization, Confounding, and Heterogeneous Effects" by Hahn, Murray and Carvalhocited by 77
  7. 2017Double/Debiased/Neyman Machine Learning of Treatment Effectscited by 76
  8. 2016Balancing Covariates via Propensity Score Weightingcited by 65
  9. 2015Nonparametric methods for doubly robust estimation of continuous treatment effectscited by 64
  10. 2018Orthogonal Random Forest for Causal Inferencecited by 51
  11. 2016Approximate Residual Balancing: De-Biased Inference of Average Treatment Effects in High Dimensionscited by 50
  12. 2017Some methods for heterogeneous treatment effect estimation in high-dimensionscited by 48
37 more
  1. 2013Robust Inference on Average Treatment Effects with Possibly More Covariates than Observationscited by 38
  2. 2020Kernel Methods for Causal Functions: Dose, Heterogeneous, and Incremental Response Curvescited by 36
  3. 2016Covariate Balancing Propensity Score by Tailored Loss Functionscited by 34
  4. 2017Augmented Minimax Linear Estimationcited by 33
  5. 2013Program Evaluation and Causal Inference with High-Dimensional Datacited by 29
  6. 2017Orthogonal Machine Learning: Power and Limitationscited by 28
  7. 2022Bayesian Causal Inference: A Critical Reviewcited by 21
  8. 2017Minimal Dispersion Approximately Balancing Weights: Asymptotic Properties and Practical Considerationscited by 18
  9. 2018Robust Inference Using Inverse Probability Weightingcited by 14
  10. 2017Selective inference for effect modification via the lassocited by 12
  11. 2018Robust Estimation of Causal Effects via High-Dimensional Covariate Balancing Propensity Scorecited by 12
  12. 2022Inference on Strongly Identified Functionals of Weakly Identified Functionscited by 12
  13. 2017Finite-Sample Optimal Estimation and Inference on Average Treatment Effects Under Unconfoundednesscited by 11
  14. 2020Debiased Inverse Propensity Score Weighting for Estimation of Average Treatment Effects with High-Dimensional Confounderscited by 11
  15. 2014Randomization Inference for Treatment Effect Variationcited by 9
  16. 2020Propensity score weighting under limited overlap and model misspecificationcited by 8
  17. 2016Doubly Robust Uniform Confidence Band for the Conditional Average Treatment Effect Functioncited by 7
  18. 2021Adaptive normalization for IPW estimationcited by 7
  19. 2015Propensity Score Matching and Subclassification in Observational Studies with Multi-level Treatmentscited by 6
  20. 2016Nonparametric Tests for Treatment Effect Heterogeneity with Duration Outcomescited by 6
  21. 2018Regularized Orthogonal Machine Learning for Nonlinear Semiparametric Modelscited by 6
  22. 2018A Causal Bootstrapcited by 4
  23. 2017Non-separable Models with High-dimensional Datacited by 3
  24. 2017A new scope of penalized empirical likelihood with high-dimensional estimating equationscited by 3
  25. 2017Estimation and Inference for Moments of Ratios with Robustness against Large Trimming Biascited by 3
  26. 2018Randomization Tests for Weak Null Hypotheses in Randomized Experimentscited by 3
  27. 2021Robust Sample Weighting to Facilitate Individualized Treatment Rule Learning for a Target Populationcited by 3
  28. 2022Overlap, matching, or entropy weights: what are we weighting for?cited by 3
  29. 2018Estimating Population Average Causal Effects in the Presence of Non-Overlap: The Effect of Natural Gas Compressor Station Exposure on Cancer Mortalitycited by 2
  30. 2018Inference under Covariate-Adaptive Randomization with Multiple Treatmentscited by 2
  31. 2020Leveraging the Fisher randomization test using confidence distributions: inference, combination and fusion learningcited by 2
  32. 2021Randomization Inference beyond the Sharp Null: Bounded Null Hypotheses and Quantiles of Individual Treatment Effectscited by 2
  33. 2021Bounding Treatment Effects by Pooling Limited Information across Observationscited by 2
  34. 2021Inference for multi-valued heterogeneous treatment effects when the number of treated units is smallcited by 1
  35. 2022What is a randomization test?cited by 1
  36. 2022Towards R-learner with Continuous Treatmentscited by 1
  37. 2022Using Balancing Weights to Target the Treatment Effect on the Treated when Overlap is Poorcited by 1

Built on by

Not cited yet.