提出新方法评估弱重叠下因果推断的可靠性
A Sensitivity Approach to Causal Inference Under Limited Overlap
- 通过最坏情况分析量化剪裁带来的偏差上限
- 在有限重叠区域给出可解释的置信区间
- 适合处理样本重叠差的观察性研究
处理组与对照组之间重叠不足是观察性研究中的关键挑战。标准方法如剪裁重要性权重虽可降低方差，但会引入根本性偏差。本文提出一种敏感性框架，用于评估在何种情况下结果函数的不规则程度会使主要结论失效。该方法基于对标准剪裁实践引入偏差的最坏情况置信界，在明确外推反事实估计所需假设的前提下进行分析。实证上，我们展示了该敏感性框架如何通过量化有限重叠区域的不确定性，防范虚假发现。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。