Matching Methods for Causal Inference: A Review and a Look Forward

Abstract

When estimating causal effects using observational data, it is desirable to replicate a randomized experiment as closely as possible by obtaining treated and control groups with similar covariate distributions. This goal can often be achieved by choosing well-matched samples of the original treated and control groups, thereby reducing bias due to the covariates. Since the 1970s, work on matching methods has examined how to best choose treated and control subjects for comparison. Matching methods are gaining popularity in fields such as economics, epidemiology, medicine and political science. However, until now the literature and related advice has been scattered across disciplines. Researchers who are interested in using matching methods---or developing methods related to matching---do not have a single place to turn to learn about past and current research. This paper provides a structure for thinking about matching methods and guidance on their use, coalescing the existing research (both old and new) and providing a summary of where the literature on matching methods is now and where it should be headed.

Published in at http://dx.doi.org/10.1214/09-STS313 the Statistical Science (http://www.imstat.org/sts/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (2)

• Demystifying Double Robustness: A Comparison of Alternative Strategies for Estimating a Population Mean from Incomplete Data
• Affinely invariant matching methods with discriminant mixtures of proportional ellipsoidally symmetric distributions

Cited by in corpus (130)

• All Models are Wrong, but Many are Useful: Learning a Variable's Importance by Studying an Entire Class of Prediction Models Simultaneously
• Valid post-selection inference
• A Review of Generalizability and Transportability
• Estimation of causal effects with multiple treatments: a review and new ideas
• Propensity score matching in SPSS
• MatchThem:: Matching and Weighting after Multiple Imputation
• Matching for balance, pairing for heterogeneity in an observational study of the effectiveness of for-profit and not-for-profit high schools in Chile
• Causally Regularized Learning with Agnostic Data Selection Bias
• Bayesian Inference of Individualized Treatment Effects using Multi-task Gaussian Processes
• Fairness in Algorithmic Decision Making: An Excursion Through the Lens of Causality
• Causal Decision Trees
• Methods for Population Adjustment with Limited Access to Individual Patient Data: A Review and Simulation Study
• Stronger instruments via integer programming in an observational study of late preterm birth outcomes
• Investigating Causality in Human Behavior from Smartphone Sensor Data: A Quasi-Experimental Approach
• Adjusting for indirectly measured confounding using large-scale propensity scores
• Automated versus do-it-yourself methods for causal inference: Lessons learned from a data analysis competition
• Parametric G-computation for Compatible Indirect Treatment Comparisons with Limited Individual Patient Data
• Scientific Prizes and the Extraordinary Growth of Scientific Topics
• Survey on Causal-based Machine Learning Fairness Notions
• A modified debiased inverse-variance weighted estimator in two-sample summary-data Mendelian randomization
• Revisiting the role of bars in AGN fuelling with propensity score sample matching
• Alpha-1 adrenergic receptor antagonists to prevent hyperinflammation and death from lower respiratory tract infection
• How to apply multiple imputation in propensity score matching with partially observed confounders: a simulation study and practical recommendations
• Selecting a Scale for Spatial Confounding Adjustment
• Assessing the causal effect of binary interventions from observational panel data with few treated units
• From Observational Studies to Causal Rule Mining
• Generalized Optimal Matching Methods for Causal Inference
• A confounding bridge approach for double negative control inference on causal effects
• A Survey on Causal Inference
• An Evaluation Toolkit to Guide Model Selection and Cohort Definition in Causal Inference
• The causal effect of a timeout at stopping an opposing run in the NBA
• Confounder selection strategies targeting stable treatment effect estimators
• Isolation in the construction of natural experiments
• Covariate Balancing Propensity Score by Tailored Loss Functions
• Non-Parametric Causality Detection: An Application to Social Media and Financial Data
• Intergroup Contact in the Wild: Characterizing Language Differences between Intergroup and Single-group Members in NBA-related Discussion Forums
• Robust policy evaluation from large-scale observational studies
• Controlled Analyses of Social Biases in Wikipedia Bios
• Quantifying How Hateful Communities Radicalize Online Users
• Online Actions with Offline Impact: How Online Social Networks Influence Online and Offline User Behavior
• Beautiful and damned. Combined effect of content quality and social ties on user engagement
• Estimating causal effects with optimization-based methods: A review and empirical comparison
• Prior-preconditioned conjugate gradient method for accelerated Gibbs sampling in "large $n$ & large $p$" Bayesian sparse regression
• Combining multiple observational data sources to estimate causal effects
• Reciprocity, Homophily, and Social Network Effects in Pictorial Communication: A Case Study of Bitmoji Stickers
• Estimating Community Feedback Effect on Topic Choice in Social Media with Predictive Modeling
• Randomization Tests that Condition on Non-Categorical Covariate Balance
• Towards Causality-Aware Inferring: A Sequential Discriminative Approach for Medical Diagnosis
• Identifying Statistical Bias in Dataset Replication
• Identifying Spurious Correlations for Robust Text Classification
• The Strength of Structural Diversity in Online Social Networks
• Electrocardiogram-based diagnosis of liver diseases: an externally validated and explainable machine learning approach
• Beyond Correlation: Incorporating Counterfactual Guidance to Better Support Exploratory Visual Analysis
• Stop the Clock: Are Timeout Effects Real?
• Causal Inference: A Missing Data Perspective
• Robust Estimation of Propensity Score Weights via Subclassification
• Uncertain Neighbors: Bayesian Propensity Score Matching For Causal Inference
• NCoRE: Neural Counterfactual Representation Learning for Combinations of Treatments
• Estimating Average Treatment Effects with Support Vector Machines
• Propensity score analysis with latent covariates: Measurement error bias correction using the covariate's posterior mean, aka the inclusive factor score
• Evaluating hybrid controls methodology in early-phase oncology trials: a simulation study based on the MORPHEUS-UC trial
• MALTS: Matching After Learning to Stretch
• Optimal Multilevel Matching in Clustered Observational Studies: A Case Study of the Effectiveness of Private Schools Under a Large-Scale Voucher System
• Prioritizing Variables for Observational Study Design using the Joint Variable Importance Plot
• Quantifying Confounding Bias in Generative Art: A Case Study
• Sensitivity analysis for principal ignorability violation in estimating complier and noncomplier average causal effects
• Bayesian propensity score matching in automotive embedded software engineering
• Adjustment for Biased Sampling Using NHANES Derived Propensity Weights
• From controlled to undisciplined data: estimating causal effects in the era of data science using a potential outcome framework
• Counterfactual Representation Learning with Balancing Weights
• Combining Offline Causal Inference and Online Bandit Learning for Data Driven Decision
• Propensity score matching for estimating a marginal hazard ratio
• MultiMBNN: Matched and Balanced Causal Inference with Neural Networks
• Randomization Inference beyond the Sharp Null: Bounded Null Hypotheses and Quantiles of Individual Treatment Effects
• Full Matching Approach to Instrumental Variables Estimation with Application to the Effect of Malaria on Stunting
• Generalized full matching and extrapolation of the results from a large-scale voter mobilization experiment
• Size matters? Or not: A/B testing with limited sample in automotive embedded software
• Linear Estimation of Treatment Effects in Demand Response: An Experimental Design Approach
• Discovering Effect Modification and Randomization Inference in Air Pollution Studies
• The Local Randomization Framework for Regression Discontinuity Designs: A Review and Some Extensions
• Bridging preference-based instrumental variable studies and cluster-randomized encouragement experiments: study design, noncompliance, and average cluster effect ratio
• Reducing bias in difference-in-differences models using entropy balancing
• Propensity Score Analysis with Matching Weights
• What can the millions of random treatments in nonexperimental data reveal about causes?
• Estimating the average treatment effects of nutritional label use using subclassification with regression adjustment
• Adaptive Hyper-box Matching for Interpretable Individualized Treatment Effect Estimation
• A calibrated sensitivity analysis for matched observational studies with application to the effect of second-hand smoke exposure on blood lead levels in U.S. children
• Interpretable Almost Matching Exactly for Causal Inference
• Generating Synthetic Text Data to Evaluate Causal Inference Methods
• A Balancing Weight Framework for Estimating the Causal Effect of General Treatments
• Causal Inference in Repeated Observational Studies: A Case Study of eBay Product Releases
• Trimmed Match Design for Randomized Paired Geo Experiments
• Split-Treatment Analysis to Rank Heterogeneous Causal Effects for Prospective Interventions
• Causal inference and machine learning approaches for evaluation of the health impacts of large-scale air quality regulations
• Social Distancing and COVID-19: Randomization Inference for a Structured Dose-Response Relationship
• Moral Judgments in Online Discourse are not Biased by Gender
• Probabilistic Matching: Causal Inference under Measurement Errors
• Detecting Heterogeneous Treatment Effect with Instrumental Variables
• Optimal Estimation of Generalized Average Treatment Effects using Kernel Optimal Matching
• Relationship between changing malaria burden and low birth weight in sub-Saharan Africa: a difference-in-differences study via a pair-of-pairs approach
• Propensity score-based estimators with multiple error-prone covariates
• Sharpening the Rosenbaum Sensitivity Bounds to Address Concerns About Interactions Between Observed and Unobserved Covariates
• Robust Bayesian causal estimation for causal inference in medical diagnosis
• Causal thinking for decision making on Electronic Health Records: why and how
• Simultaneous causal inference for multiple treatments via sufficiency
• Shift in house price estimates during COVID-19 reveals effect of crisis on collective speculation
• A Menu-Driven Software Package of Bayesian Nonparametric (and Parametric) Mixed Models for Regression Analysis and Density Estimation
• Matching methods for obtaining survival functions to estimate the effect of a time-dependent treatment
• Functional Data Analysis with Causation in Observational Studies: Covariate Balancing Functional Propensity Score for Functional Treatments
• Algorithms and Complexity for Variants of Covariates Fine Balance
• Weighting-Based Treatment Effect Estimation via Distribution Learning
• Protocol for an Observational Study of the Association of High School Football Participation on Health in Late Adulthood
• ZaliQL: A SQL-Based Framework for Drawing Causal Inference from Big Data
• A comparison of methods for estimating the average treatment effect on the treated for externally controlled trials
• Re-Evaluating Strengthened-IV Designs: Asymptotic Efficiency, Bias Formula, and the Validity and Power of Sensitivity Analyses
• Causal Inference for Quantile Treatment Effects
• Text as Causal Mediators: Research Design for Causal Estimates of Differential Treatment of Social Groups via Language Aspects
• Resource Allocation in One-dimensional Distributed Service Networks with Applications
• $\beta$-Intact-VAE: Identifying and Estimating Causal Effects under Limited Overlap
• Conditional As-If Analyses in Randomized Experiments
• Discovering Context Specific Causal Relationships
• Ridge Rerandomization: An Experimental Design Strategy in the Presence of Collinearity
• A deterministic matching method for exact matchings to compare the outcome of different interventions
• Marginalization of Regression-Adjusted Treatment Effects in Indirect Comparisons with Limited Patient-Level Data
• Addressing Spatially Structured Interference in Causal Analysis Using Propensity Scores
• Comparing Broadband ISP Performance using Big Data from M-Lab
• Causality-Aware Neighborhood Methods for Recommender Systems
• Optimal transport weights for causal inference
• Matched sample selection with GANs for mitigating attribute confounding
• Efficient Balanced Treatment Assignments for Experimentation