A Sensitivity Approach to Causal Inference Under Limited Overlap

TL;DR

This paper introduces a sensitivity framework for causal inference in observational studies with limited overlap, assessing how outcome irregularities can invalidate findings.

Contribution

It develops a worst-case bias bounding method to contextualize causal estimates under limited overlap conditions.

Findings

Framework quantifies uncertainty in regions with limited overlap.

Method protects against spurious causal findings.

Empirical demonstration shows robustness of the approach.

Abstract

Limited overlap between treated and control groups is a key challenge in observational analysis. Standard approaches like trimming importance weights can reduce variance but introduce a fundamental bias. We propose a sensitivity framework for contextualizing findings under limited overlap, where we assess how irregular the outcome function has to be in order for the main finding to be invalidated. Our approach is based on worst-case confidence bounds on the bias introduced by standard trimming practices, under explicit assumptions necessary to extrapolate counterfactual estimates from regions of overlap to those without. Empirically, we demonstrate how our sensitivity framework protects against spurious findings by quantifying uncertainty in regions with limited overlap.