A Sensitivity Approach to Causal Inference Under Limited Overlap
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917416623144960 |
|---|---|
| author | Ma, Yuanzhe Huang, Yian Namkoong, Hongseok |
| author_facet | Ma, Yuanzhe Huang, Yian Namkoong, Hongseok |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_22003 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Sensitivity Approach to Causal Inference Under Limited Overlap Ma, Yuanzhe Huang, Yian Namkoong, Hongseok Machine Learning Methodology 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. |
| title | A Sensitivity Approach to Causal Inference Under Limited Overlap |
| topic | Machine Learning Methodology |
| url | https://arxiv.org/abs/2511.22003 |