Corpus ID: 33348619

A Framework for Optimal Matching for Causal Inference

  title={A Framework for Optimal Matching for Causal Inference},
  author={Nathan Kallus},
  • Nathan Kallus
  • Published in AISTATS 2017
  • Mathematics, Computer Science
  • We propose a novel framework for matching estimators for causal effect from observational data that is based on minimizing the dual norm of estimation error when expressed as an operator. We show that many popular matching estimators can be expressed as optimal in this framework, including nearest-neighbor matching, coarsened exact matching, and mean-matched sampling. This reveals their motivation and aptness as structural priors formulated by embedding the effect in a particular functional… CONTINUE READING
    18 Citations
    Generalized Optimal Matching Methods for Causal Inference
    • Nathan Kallus
    • Computer Science, Mathematics
    • J. Mach. Learn. Res.
    • 2020
    • 39
    • PDF
    MALTS: Matching After Learning to Stretch
    • 9
    • PDF
    A Balancing Weight Framework for Estimating the Causal Effect of General Treatments.
    A Causal Dirichlet Mixture Model for Causal Inference from Observational Data
    The Counterfactual χ-GAN
    Doubly robust off-policy evaluation with shrinkage
    • 1
    Doubly robust off-policy evaluation with shrinkage
    • 8
    • PDF
    The Counterfactual $\chi$-GAN
    Policy Evaluation with Latent Confounders via Optimal Balance
    • 6
    • PDF


    A Theory of Statistical Inference for Matching Methods in Applied Causal Research ∗
    • 10
    • PDF
    Bayesian Nonparametric Modeling for Causal Inference
    • 470
    • Highly Influential
    Matching methods for causal inference: A review and a look forward.
    • E. Stuart
    • Computer Science, Medicine
    • Statistical science : a review journal of the Institute of Mathematical Statistics
    • 2010
    • 2,530
    • PDF
    Optimal Matching for Observational Studies
    • 516
    • Highly Influential
    • PDF
    Optimal A Priori Balance in the Design of Controlled Experiments
    • 48
    • PDF
    Causal Inference without Balance Checking: Coarsened Exact Matching
    • 1,520
    • Highly Influential
    • PDF
    Why Propensity Scores Should Not Be Used for Matching
    • 391
    • PDF
    A Kernel Method for the Two-Sample-Problem
    • 1,213
    • PDF