Decomposing Treatment Effect Variation
Decomposing Treatment Effect Variation
复制标题
DOI：
10.1080/01621459.2017.1407322
复制
发表时间：
2019-01-02
影响因子：
3.7
通讯作者：
Miratrix, Luke
中科院分区：
文献类型：
--
作者：
Ding, Peng;Feller, Avi;Miratrix, Luke
理解和表征随机实验中的治疗效果变化对于超越平均治疗效果的黑箱已经变得至关重要。然而，传统的统计方法往往忽略或假设了这种变化。在随机化实验的背景下，本文提出了一个框架，用于将整体治疗效果变化分解为由观察到的协变量解释的系统性成分和剩余的特异质成分。我们的框架是完全基于随机化的，治疗效果变化的估计完全由随机化本身证明。我们的框架也可以解释不遵守的情况，这是一个重要的实际复杂性。我们做出了一些贡献。首先，我们表明，基于随机化的系统变异估计是非常相似的形式，从充分相互作用的线性回归和两阶段最小二乘估计。其次，我们使用这些估计量来开发一个综合测试系统的治疗效果的变化，无论有没有不遵守。第三，我们提出了一个类似R-2的测量协变量解释的治疗效果的变化，并在适用的情况下，不遵守。最后，我们通过模拟研究评估这些方法，并将其应用于领先影响研究，一个大规模的随机实验。本文的补充材料可在网上查阅。
Understanding and characterizing treatment effect variation in randomized experiments has become essential for going beyond the black box of the average treatment effect. Nonetheless, traditional statistical approaches often ignore or assume away such variation. In the context of randomized experiments, this article proposes a framework for decomposing overall treatment effect variation into a systematic component explained by observed covariates and a remaining idiosyncratic component. Our framework is fully randomization-based, with estimates of treatment effect variation that are entirely justified by the randomization itself. Our framework can also account for noncompliance, which is an important practical complication. We make several contributions. First, we show that randomization-based estimates of systematic variation are very similar in form to estimates from fully interacted linear regression and two-stage least squares. Second, we use these estimators to develop an omnibus test for systematic treatment effect variation, both with and without noncompliance. Third, we propose an R-2-like measure of treatment effect variation explained by covariates and, when applicable, noncompliance. Finally, we assess these methods via simulation studies and apply them to the Head Start Impact Study, a large-scale randomized experiment. Supplementary materials for this article are available online.