Addressing Extreme Propensity Scores via the Overlap Weights
Addressing Extreme Propensity Scores via the Overlap Weights
复制标题
DOI：
10.1093/aje/kwy201
复制
发表时间：
2019-01-01
影响因子：
5
通讯作者：
Li, Fan
中科院分区：
文献类型：
--
作者：
Li, Fan;Thomas, Laine E.;Li, Fan
因果推理中流行的逆概率加权方法常常受到极端倾向得分的阻碍，导致估计有偏差和方差过大。常见的补救措施是修剪得分极端的患者（即将他们从加权分析中删除）。然而，此类方法通常对截止点的选择很敏感，并且会丢弃很大一部分样本。治疗效果估计的偏倚和精确度的影响尚不清楚。新开发的方法——重叠加权法可以缓解这些问题。重叠权重通过不断降低倾向得分分布尾部单位的权重，强调治疗之间观察到的特征重叠最多的目标人群。在这里，我们使用模拟将重叠权重与带有修剪的标准逆概率加权在偏差、方差和 95% 置信区间覆盖率方面进行比较。考虑了一系列倾向得分分布，包括具有大量非重叠和极值的设置。为了便于实际实施，我们进一步为使用重叠加权估计的治疗效果的标准误差提供了一致的估计器。
The popular inverse probability weighting method in causal inference is often hampered by extreme propensity scores, resulting in biased estimates and excessive variance. A common remedy is to trim patients with extreme scores (i.e., remove them from the weighted analysis). However, such methods are often sensitive to the choice of cutoff points and discard a large proportion of the sample. The implications for bias and the precision of the treatment effect estimate are unclear. These problems are mitigated by a newly developed method, the overlap weighting method. Overlap weights emphasize the target population with the most overlap in observed characteristics between treatments, by continuously down-weighting the units in the tails of the propensity score distribution. Here we use simulations to compare overlap weights to standard inverse probability weighting with trimming, in terms of bias, variance, and 95% confidence interval coverage. A range of propensity score distributions are considered, including settings with substantial nonoverlap and extreme values. To facilitate practical implementation, we further provide a consistent estimator for the standard error of the treatment effect estimated using overlap weighting.