Quasi-oracle estimation of heterogeneous treatment effects
2020/09/09 by X Nie, Xinkun Nie, Stefan Wager +1 · 120 citations
Mathematics · #A priori and a posteriori #Advanced Causal Inference Techniques #Artificial intelligence #Boosting (machine learning) #Computer science #Estimator #Kernel (algebra) #Kernel regression #Machine learning #Mathematical optimization #Mathematics #Oracle #Regression #Statistical Methods and Inference #Statistical Methods in Clinical Trials #Statistics
paper · doi:10.1093/biomet/asaa076
published in Biometrika 108(2), 299-319 (Oxford University Press)
openalex publication_date 2020/09/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/15
Abstract
Summary Flexible estimation of heterogeneous treatment effects lies at the heart of many statistical applications, such as personalized medicine and optimal resource allocation. In this article we develop a general class of two-step algorithms for heterogeneous treatment effect estimation in observational studies. First, we estimate marginal effects and treatment propensities to form an objective function that isolates the causal component of the signal. Then, we optimize this data-adaptive objective function. The proposed approach has several advantages over existing methods. From a practical perspective, our method is flexible and easy to use: in both steps, any loss-minimization method can be employed, such as penalized regression, deep neural networks, or boosting; moreover, these methods can be fine-tuned by cross-validation. Meanwhile, in the case of penalized kernel regression, we show that our method has a quasi-oracle property. Even when the pilot estimates for marginal effects and treatment propensities are not particularly accurate, we achieve the same error bounds as an oracle with prior knowledge of these two nuisance components. We implement variants of our approach based on penalized regression, kernel ridge regression, and boosting in a variety of simulation set-ups, and observe promising performance relative to existing baselines.
Citations
Cited by
- Interventional Score Geometry for Causal Inference
- Learning Who to Treat When Treatment is Missing
- A Variance-Based Test for Heterogeneous Treatment Effects
- Generative Synthetic Data for Causal Inference: Pitfalls, Remedies, and Opportunities
- Counterfactual Optimal Action Trees (COAT): Interpretable Prescriptive Policies from Observational Data
- Bi-Level Decision-Focused Causal Learning for Large-Scale Marketing Optimization: Bridging Observational and Experimental Data
- Propensity Patchwork Kriging for Scalable Inference on Heterogeneous Treatment Effects
- TLRNet: Estimating Individual Treatment Effect based on Local Information and Single Learner Structure
- PIPCFR: Pseudo-outcome Imputation with Post-treatment Variables for Individual Treatment Effect Estimation
- Policy-Aligned Estimation of Conditional Average Treatment Effects
- Supervised learning pays attention
- Non-parametric assessment of the calibration of individualized treatment effects
- Towards a Relationship-Aware Transformer for Tabular Data
- A General Bayesian Nonparametric Approach for Estimating Population-Level and Conditional Causal Effects
- The Final-Stage Bottleneck: A Systematic Dissection of the R-Learner for Network Causal Inference
- A Large Scale Heterogeneous Treatment Effect Estimation Framework and Its Applications of Users' Journey at Snap
- Learning Subgroups with Maximum Treatment Effects without Causal Heuristics
- Reliable Selection of Heterogeneous Treatment Effect Estimators
- A Comparative Study of Model Adaptation Strategies for Multi-Treatment Uplift Modeling
- Assessment of the conditional exchangeability assumption in causal machine learning models: a simulation study
- From Unsupervised Subgroups to Hypothetical State-Intervention Policies: An Evaluation of Selected Subgrouping Methods in Observational Health Data
- Uncertainty-Guided LLM Semantic Augmentation for Heterogeneous Treatment Effect Estimation
- TERRA: A Transformer-Enabled Recursive R-learner for Longitudinal Heterogeneous Treatment Effect Estimation
- Leveraging semantic similarity for experimentation with AI-generated treatments
- Overlap-weighted orthogonal meta-learner for treatment effect estimation over time
- A New Targeted-Federated Learning Framework for Estimating Heterogeneity of Treatment Effects: A Robust Framework with Applications in Aging Cohorts
- Estimation of causal dose-response functions under data fusion
- Inference on Variable Importance for Treatment Effect Heterogeneity: Shapley Values and Beyond
- Assessing the robustness of heterogeneous treatment effects in survival analysis under informative censoring
- Denoised IPW-Lasso for Heterogeneous Treatment Effect Estimation in Randomized Experiments
- An Orthogonal Learner for Individualized Outcomes in Markov Decision Processes
- Overlap-Adaptive Regularization for Conditional Average Treatment Effect Estimation
- Estimating and discovering heterogeneous treatment effects using machine learning in epidemiological studies: a practical guide
- Meta-Router: Bridging Gold-standard and Preference-based Evaluations in Large Language Model Routing
- GDR-learners: Orthogonal Learning of Generative Models for Potential Outcomes
- CausalKANs: interpretable treatment effect estimation with Kolmogorov-Arnold networks
- Debiased Front-Door Learners for Heterogeneous Effects
- Multi-channel Uplift Policy Learning
- Estimating Treatment Effects for Depression in Longitudinal Therapy Switching Settings
- Estimating Heterogeneous Causal Effect on Networks via Orthogonal Learning
- DoubleGen: Debiased Generative Modeling of Counterfactuals
- Cumulative Marginal Mean Model for Assessing Sequential Effects Using Digital Health Data
- Decision Theoretic Subgroup Detection With Bayesian Machine Learning
- Causal Clustering for Conditional Average Treatment Effects Estimation and Subgroup Discovery
- Precision Mental Health: Predicting Heterogeneous Treatment Effects for Depression through Data Integration
- Matching-Based Nonparametric Estimation of Group Average Treatment Effects
- Counterfactual Probabilistic Diffusion with Expert Models
- CTRL Your Shift: Clustered Transfer Residual Learning for Many Small Datasets
- ADT4Coupons: An Innovative Framework for Sequential Coupon Distribution in E-commerce
- Optimal Transport Learning: Balancing Value Optimization and Fairness in Individualized Treatment Rules
- Annotation-Assisted Learning of Treatment Policies From Multimodal Electronic Health Records
- Cross-World Assumption and Refining Prediction Intervals for Individual Treatment Effects
- An Algorithm for Identifying Interpretable Subgroups With Elevated Treatment Effects
- Bridging Structural Causal Inference and Machine Learning The S-DIDML Estimator for Heterogeneous Treatment Effects
- Late Fusion Multi-task Learning for Semiparametric Inference with Nuisance Parameters
- The Debiased Score Test: Hunt-and-test for Semiparametric Hypotheses
- Deep Learning of Continuous and Structured Policies for Aggregated Heterogeneous Treatment Effects
- Heterogeneous Causal Learning for Optimizing Aggregated Functions in User Growth
- Consistent Labeling Across Group Assignments: Variance Reduction in Conditional Average Treatment Effect Estimation
- LILI clustering algorithm: Limit Inferior Leaf Interval Integrated into Causal Forest for Causal Interference
- Robust estimation of heterogeneous treatment effects in randomized trials leveraging external data
- LLM-Driven Treatment Effect Estimation Under Inference Time Text Confounding
- Shrinkage-Based Regressions with Many Related Treatments
- Optimal Targeting in Dynamic Systems
- Learning Causally Predictable Outcomes from Psychiatric Longitudinal Data
- Finite-sample bias-variance tradeoff with variables related to trial participation inserted into causal forest models for ensuring generalizability
- Foundation Models for Causal Inference via Prior-Data Fitted Networks
- Do-PFN: In-Context Learning for Causal Effect Estimation
- On Efficient Estimation of Distributional Treatment Effects under Covariate-Adaptive Randomization
- A Diffusion-Based Method for Learning the Multi-Outcome Distribution of Medical Treatments
- Doubly Robust Alignment for Large Language Models
- Estimating Misreporting in the Presence of Genuine Modification: A Causal Perspective
- TabPFN: One Model to Rule Them All?
- M-learner:A Flexible And Powerful Framework To Study Heterogeneous Treatment Effect In Mediation Model
- A Distributionally Robust Framework for Nuisance in Causal Effect Estimation
- UpliftBench: Revealing Outcome-Regime and Objective Mismatch in Uplift Evaluation
- Treatment Effect Estimation for Optimal Decision-Making
- Orthogonal Survival Learners for Estimating Heterogeneous Treatment Effects from Time-to-Event Data
- Varying impacts of letters of recommendation on college admissions: Approximate balancing weights for subgroup effects in observational studies
- Reinforcement Learning for Individual Optimal Policy from Heterogeneous Data
- Federated Causal Inference in Healthcare: Methods, Challenges, and Applications
- Overview and practical recommendations on using Shapley Values for identifying predictive biomarkers via CATE modeling
- Geometric Causal Models
- Rank-Learner: Orthogonal Ranking of Treatment Effects
- ORTHOBO: Orthogonal Bayesian Hyperparameter Optimization
- Statistical Learning for Heterogeneous Treatment Effects: Pretraining, Prognosis, and Prediction
- Generalized Bayes for Causal Inference
- Detecting and Explaining Unlawful Insider Trading: A Shapley Value and Causal Forest Approach to Identifying Key Drivers and Causal Relationships
- Beyond the Numbers: Causal Effects of Financial Report Sentiment on Bank Profitability
- Adaptive sample splitting for randomization tests
- MOSIC: Model-Agnostic Optimal Subgroup Identification with Multi-Constraint for Improved Reliability
- Observationally Informed Adaptive Causal Experimental Design
- Selective randomization inference for subgroup effects with continuous biomarkers
- Causal rule ensemble approach for multi-arm data
- Dynamic Regularized CBDT: Variance-Calibrated Causal Boosting for Interpretable Heterogeneous Treatment Effects
- A conceptual synthesis of causal assumptions for causal discovery and inference
- Heterogeneous Effects of Health Insurance on Rural Children's Health in China: A Causal Machine Learning Approach. [europepmc]
- Beyond the Mean: A Flexible Framework for Studying Causal Effects Using Linear Models. [europepmc]
- The effect of COVID certificates on vaccine uptake, health outcomes, and the economy. [europepmc]
- Learning Causal Effects From Observational Data in Healthcare: A Review and Summary. [europepmc]
- A reference-free R-learner for treatment recommendation. [europepmc]
- How causal machine learning can leverage marketing strategies: Assessing and improving the performance of a coupon campaign. [europepmc]
- The effect of COVID-19 on self-reported safety incidents in aviation: An examination of the heterogeneous effects using causal machine learning. [europepmc]
- Effect of Angiotensin-Converting Enzyme Inhibitor and Angiotensin Receptor Blocker Initiation on Organ Support-Free Days in Patients Hospitalized With COVID-19: A Randomized Clinical Trial. [europepmc]
- Updated beliefs and shaken confidence: evidence from vaccine hesitancy caused by experiencing "COVID arm". [europepmc]
- Estimated Average Treatment Effect of Psychiatric Hospitalization in Patients With Suicidal Behaviors: A Precision Treatment Analysis. [europepmc]
- BOUNDS ON THE CONDITIONAL AND AVERAGE TREATMENT EFFECT WITH UNOBSERVED CONFOUNDING FACTORS. [europepmc]
- Comparison of methods that combine multiple randomized trials to estimate heterogeneous treatment effects. [europepmc]
- Estimating individualized treatment effects using an individual participant data meta-analysis. [europepmc]
- Heterogeneity of the effect of the COVID-19 pandemic on the incidence of Metabolic Syndrome onset at a Japanese campus. [europepmc]
- Methods for Integrating Trials and Non-experimental Data to Examine Treatment Effect Heterogeneity. [europepmc]
- Design of a multicenter randomized controlled trial of a post-discharge suicide prevention intervention for high-risk psychiatric inpatients: The Veterans Coordinated Community Care Study. [europepmc]
- A RELUCTANT ADDITIVE MODEL FRAMEWORK FOR INTERPRETABLE NONLINEAR INDIVIDUALIZED TREATMENT RULES. [europepmc]
- Evaluating Meta-Learners to Analyze Treatment Heterogeneity in Survival Data: Application to Electronic Health Records of Pediatric Asthma Care in COVID-19 Pandemic. [europepmc]
- Two-step pragmatic subgroup discovery for heterogeneous treatment effects analyses: perspectives toward enhanced interpretability. [europepmc]
- An overview of modern machine learning methods for effect measure modification analyses in high-dimensional settings. [europepmc]
- Estimating Treatment Effect Heterogeneity in Psychiatry: A Review and Tutorial With Causal Forests. [europepmc]
- Doubly robust machine learning-based estimation methods for instrumental variables with an application to surgical care for cholecystitis. [europepmc]
- Causal inference unveils how forest coverage mitigates excess snakebite cases during rainfall seasons in Colombia. [europepmc]
- Estimating causal effects with optimization-based methods: A review and empirical comparison. [europepmc]
- The effect of COVID certificates on vaccine uptake, health outcomes, and the economy [europepmc]
Related