Interpretable Almost-Matching-Exactly With Instrumental Variables

TL;DR

This paper introduces an interpretable matching framework for instrumental variables that improves causal effect estimation in observational studies by addressing unmeasured confounding with better matching algorithms.

Contribution

It proposes a novel matching approach for IV that handles categorical confounders, scales well, and outperforms existing methods in simulations and real-world application.

Findings

Better matching performance than existing methods on simulated data

Effective in reducing unmeasured confounding in causal inference

Provides interpretable matches for observational data analysis

Abstract

Uncertainty in the estimation of the causal effect in observational studies is often due to unmeasured confounding, i.e., the presence of unobserved covariates linking treatments and outcomes. Instrumental Variables (IV) are commonly used to reduce the effects of unmeasured confounding. Existing methods for IV estimation either require strong parametric assumptions, use arbitrary distance metrics, or do not scale well to large datasets. We propose a matching framework for IV in the presence of observed categorical confounders that addresses these weaknesses. Our method first matches units exactly, and then consecutively drops variables to approximately match the remaining units on as many variables as possible. We show that our algorithm constructs better matches than other existing methods on simulated datasets, and we produce interesting results in an application to political canvassing.