COVARIATE BALANCING PROPENSITY SCORE FOR A CONTINUOUS TREATMENT: APPLICATION TO THE EFFICACY OF POLITICAL ADVERTISEMENTS
COVARIATE BALANCING PROPENSITY SCORE FOR A CONTINUOUS TREATMENT: APPLICATION TO THE EFFICACY OF POLITICAL ADVERTISEMENTS
复制标题
DOI：
10.1214/17-aoas1101
复制
发表时间：
2018-03-01
影响因子：
1.8
通讯作者：
Imai, Kosuke
中科院分区：
文献类型：
--
作者：
Fong, Christian;Hazlett, Chad;Imai, Kosuke
在观察性研究中，倾向评分匹配和加权是估计因果效应的常用方法。然而，除了不混淆的假设之外，这些方法还需要正确指定倾向评分的模型。最近提出的协变量平衡倾向评分（CBPS）方法通过直接优化治疗组和对照组之间的样本协变量平衡来增加对模型错误指定的稳健性。在本文中，我们将CBPS扩展到一个连续的治疗。我们提出了协变量平衡广义倾向评分（CBGPS）方法，该方法最大限度地减少了协变量和治疗之间的关联。我们开发参数和非参数的方法，并显示其上级性能优于标准的最大似然估计在模拟研究。CBGPS方法应用于观察性研究，其目标是估计竞选捐款的政治广告的因果关系。我们还提供了实现所提出的方法的开源软件。
Propensity score matching and weighting are popular methods when estimating causal effects in observational studies. Beyond the assumption of unconfoundedness, however, these methods also require the model for the propensity score to be correctly specified. The recently proposed covariate balancing propensity score (CBPS) methodology increases the robustness to model misspecification by directly optimizing sample covariate balance between the treatment and control groups. In this paper, we extend the CBPS to a continuous treatment. We propose the covariate balancing generalized propensity score (CBGPS) methodology, which minimizes the association between covariates and the treatment. We develop both parametric and non-parametric approaches and show their superior performance over the standard maximum likelihood estimation in a simulation study. The CBGPS methodology is applied to an observational study, whose goal is to estimate the causal effects of political advertisements on campaign contributions. We also provide open-source software that implements the proposed methods.