A Sensitivity Approach to Causal Inference Under Limited Overlap
By: Yuanzhe Ma, Hongseok Namkoong
Potential Business Impact:
Finds truth when data is incomplete.
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.
Similar Papers
A Sensitivity Analysis Framework for Causal Inference Under Interference
Methodology
Finds hidden problems affecting results.
Robust Sensitivity Analysis via Augmented Percentile Bootstrap under Simultaneous Violations of Unconfoundedness and Overlap
Methodology
Finds hidden causes even with bad data.
Controlling the False Discovery Proportion in Matched Observational Studies
Methodology
Find real treatment effects even with hidden bias.