Recursive partitioning for heterogeneous causal effects
From MaRDI portal
Publication:2962333
DOI10.1073/PNAS.1510489113zbMATH Open1357.62190arXiv1504.01132OpenAlexW2305754340WikidataQ27320968 ScholiaQ27320968MaRDI QIDQ2962333FDOQ2962333
Authors: Susan Athey, Guido W. Imbens 
Publication date: 16 February 2017
Published in: Proceedings of the National Academy of Sciences (Search for Journal in Brave)
Abstract: In this paper we study the problems of estimating heterogeneity in causal effects in experimental or observational studies and conducting inference about the magnitude of the differences in treatment effects across subsets of the population. In applications, our method provides a data-driven approach to determine which subpopulations have large or small treatment effects and to test hypotheses about the differences in these effects. For experiments, our method allows researchers to identify heterogeneity in treatment effects that was not specified in a pre-analysis plan, without concern about invalidating inference due to multiple testing. In most of the literature on supervised machine learning (e.g. regression trees, random forests, LASSO, etc.), the goal is to build a model of the relationship between a unit's attributes and an observed outcome. A prominent role in these methods is played by cross-validation which compares predictions to actual outcomes in test samples, in order to select the level of complexity of the model that provides the best predictive power. Our method is closely related, but it differs in that it is tailored for predicting causal effects of a treatment rather than a unit's outcome. The challenge is that the "ground truth" for a causal effect is not observed for any individual unit: we observe the unit with the treatment, or without the treatment, but not both at the same time. Thus, it is not obvious how to use cross-validation to determine whether a causal effect has been accurately predicted. We propose several novel cross-validation criteria for this problem and demonstrate through simulations the conditions under which they perform better than standard methods for the problem of causal effects. We then apply the method to a large-scale field experiment re-ranking results on a search engine.
Full work available at URL: https://arxiv.org/abs/1504.01132
Recommendations
- Estimation and Inference of Heterogeneous Treatment Effects using Random Forests
- Bayesian regression tree models for causal inference: regularization, confounding, and heterogeneous effects (with discussion)
- Causal interaction trees: Finding subgroups with heterogeneous treatment effects in observational data
- High-dimensional regression adjustments in randomized experiments
- Estimating treatment effect heterogeneity in randomized program evaluation
causal inferencecross-validationpotential outcomesheterogeneous treatment effectssupervised machine learning
Cites Work
- Estimating treatment effect heterogeneity in randomized program evaluation
- Title not available (Why is that?)
- Title not available (Why is that?)
- Random forests
- Estimation and Inference of Heterogeneous Treatment Effects using Random Forests
- A simple method for estimating interactions between a treatment and a large number of covariates
- Causal inference for statistics, social, and biomedical sciences. An introduction
- Observational studies.
- The central role of the propensity score in observational studies for causal effects
- Statistics and Causal Inference
- Title not available (Why is that?)
- Efficient Estimation of Average Treatment Effects Using the Estimated Propensity Score
- Large Sample Properties of Matching Estimators for Average Treatment Effects
- Title not available (Why is that?)
- Bayesian inference for causal effects: The role of randomization
- Optimizing randomized trial designs to distinguish which subpopulations benefit from treatment
Cited In (92)
- Counterfactual explanation of machine learning survival models
- Robust inference of conditional average treatment effects using dimension reduction
- Causal inference: a missing data perspective
- Subgroup causal effect identification and estimation via matching tree
- Response transformation and profit decomposition for revenue uplift modeling
- Causal interaction trees: Finding subgroups with heterogeneous treatment effects in observational data
- Stable Discovery of Interpretable Subgroups via Calibration in Causal Studies
- Comments on: ``A random forest guided tour
- Targeted smooth Bayesian causal forests: an analysis of heterogeneous treatment effects for simultaneous vs. interval medical abortion regimens over gestation
- Bayesian regression tree models for causal inference: regularization, confounding, and heterogeneous effects (with discussion)
- High-Dimensional Precision Medicine From Patient-Derived Xenografts
- Interval Censored Recursive Forests
- Non-separable models with high-dimensional data
- Learning causal effect using machine learning with application to China's typhoon
- On the trade-off between number of examples and precision of supervision in machine learning problems
- A Penalized Synthetic Control Estimator for Disaggregated Data
- How R Helps Airbnb Make the Most of its Data
- Estimation and validation of ratio-based conditional average treatment effects using observational data
- Sufficient dimension reduction for average causal effect estimation
- Inference on heterogeneous treatment effects in high‐dimensional dynamic panels under weak dependence
- Detecting heterogeneous treatment effects with instrumental variables and application to the Oregon Health Insurance Experiment
- Causal inference: Critical developments, past and future
- Automated versus do-it-yourself methods for causal inference: lessons learned from a data analysis competition
- Title not available (Why is that?)
- Estimation and Inference of Heterogeneous Treatment Effects using Random Forests
- Towards optimal doubly robust estimation of heterogeneous causal effects
- TESTING FOR UNOBSERVED HETEROGENEOUS TREATMENT EFFECTS WITH OBSERVATIONAL DATA
- Estimating treatment effect heterogeneity in randomized program evaluation
- Semiparametric Bayesian causal inference
- HETEROGENEOUS TREATMENT EFFECTS OF NUDGE AND REBATE: CAUSAL MACHINE LEARNING IN A FIELD EXPERIMENT ON ELECTRICITY CONSERVATION
- Causal interaction in factorial experiments: application to conjoint analysis
- Heterogeneous causal effects with imperfect compliance: a Bayesian machine learning approach
- Estimating causal effects with optimization-based methods: a review and empirical comparison
- Decomposing Treatment Effect Variation
- Generalized random forests
- Joint and marginal causal effects for binary non-independent outcomes
- Analysing a built-in advantage in asymmetric darts contests using causal machine learning
- The designed bootstrap for causal inference in big observational data
- Continuous treatment effect estimation via generative adversarial de-confounding
- Title not available (Why is that?)
- Discovering heterogeneous exposure effects using randomization inference in air pollution studies
- Augmented direct learning for conditional average treatment effect estimation with double robustness
- Optimal data collection design in machine learning: the case of the fixed effects generalized least squares panel data model
- Experimental Evaluation of Individualized Treatment Rules
- Optimal trade-off between sample size, precision of supervision, and selection probabilities for the unbalanced fixed effects panel data model
- Shrinkage Bayesian Causal Forests for Heterogeneous Treatment Effects Estimation
- Bounds on the conditional and average treatment effect with unobserved confounding factors
- Beyond the mean: a flexible framework for studying causal effects using linear models
- Robust machine learning for treatment effects in multilevel observational studies under cluster-level unmeasured confounding
- Constructing effective personalized policies using counterfactual inference from biased data sets with many features
- Local Linear Forests
- Recursive partitioning and multi-scale modeling on conditional densities
- Using knockoffs for controlled predictive biomarker identification
- Heterogeneous treatment effect-based random forest: HTERF
- Instrument Validity Tests With Causal Forests
- Improved inference for doubly robust estimators of heterogeneous treatment effects
- Debiasing SHAP scores in random forests
- Minimax rates for heterogeneous causal effect estimation
- Subgroup analysis and adaptive experiments crave for debiasing
- Robust estimation of heterogeneous treatment effects: an algorithm-based approach
- The Effect of Job Loss and Unemployment Insurance on Crime in Brazil
- When causality meets fairness: a survey
- Evaluating the predictive performance of subtyping: a criterion for cluster mean-based prediction
- Learning and confirming a class of treatment responders in clinical trials
- Multi-threshold proportional hazards model and subgroup identification
- Comparing algorithms for characterizing treatment effect heterogeneity in randomized trials
- Estimating individual treatment effects by gradient boosting trees
- Rule ensemble method with adaptive group Lasso for heterogeneous treatment effect estimation
- Exploratory identification of predictive biomarkers in randomized trials with normal endpoints
- Causal mediation analysis with latent subgroups
- Assessment of heterogeneous treatment effect estimation accuracy via matching
- Bayesian graphical modeling for heterogeneous causal effects
- Toward Optimal Variance Reduction in Online Controlled Experiments
- Building Trees for Probabilistic Prediction via Scoring Rules
- Exploring uplift modeling with high class imbalance
- The application of Bayesian method in estimating heterogeneity of treatment effect
- A more credible approach to parallel trends
- A welfare analysis of occupational licensing in U.S. states
- Hazed and confused: the effect of air pollution on dementia
- IQ, expectations, and choice
- Optimal feedback in contests
- Save, spend, or give? A model of housing, family insurance, and savings in old age
- Stratification trees for adaptive randomisation in randomised controlled trials
- Testing the production approach to markup estimation
- Unemployment insurance in macroeconomic stabilization
- Evaluating individualized treatment effect predictions: a model-based perspective on discrimination and calibration assessment
- Modern approaches for evaluating treatment effect heterogeneity from clinical trials and observational data
- Is there a role for statistics in artificial intelligence?
- A comparison of resampling and recursive partitioning methods in random forest for estimating the asymptotic variance using the infinitesimal jackknife
- Matching Using Sufficient Dimension Reduction for Causal Inference
- Constrained optimization for stratified treatment rules with multiple responses of survival data
- Composite interaction tree for simultaneous learning of optimal individualized treatment rules and subgroups
Uses Software
This page was built for publication: Recursive partitioning for heterogeneous causal effects
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2962333)
