Estimation and Inference of Heterogeneous Treatment Effects using Random Forests
Estimation and Inference of Heterogeneous Treatment Effects using Random Forests
复制标题
DOI：
10.1080/01621459.2017.1319839
复制
发表时间：
2018-01-01
影响因子：
3.7
通讯作者：
Athey, Susan
中科院分区：
文献类型：
--
作者：
Wager, Stefan;Athey, Susan
许多科学和工程挑战，从个性化医疗到定制营销方案，都需要了解治疗效果的异质性。在这篇文章中，我们开发了一个非参数因果森林估计异质性的治疗效果，扩展Breiman的广泛使用的随机森林算法。在潜在的结果框架与unconfoundedness，我们表明，因果森林是逐点一致的真正的治疗效果，并有一个渐近高斯和中心抽样分布。我们还讨论了一个实用的方法，用于构建渐近置信区间的真正的治疗效果，集中在因果森林估计。我们的理论结果依赖于一个通用的高斯理论的一个大家庭的随机森林算法。据我们所知，这是第一组允许任何类型的随机森林（包括分类和回归森林）用于可证明有效的统计推断的结果。在实验中，我们发现因果森林比基于最近邻匹配的经典方法更强大，特别是在存在不相关协变量的情况下。
Many scientific and engineering challengesranging from personalized medicine to customized marketing recommendationsrequire an understanding of treatment effect heterogeneity. In this article, we develop a nonparametric causal forest for estimating heterogeneous treatment effects that extends Breiman's widely used random forest algorithm. In the potential outcomes framework with unconfoundedness, we show that causal forests are pointwise consistent for the true treatment effect and have an asymptotically Gaussian and centered sampling distribution. We also discuss a practical method for constructing asymptotic confidence intervals for the true treatment effect that are centered at the causal forest estimates. Our theoretical results rely on a generic Gaussian theory for a large family of random forest algorithms. To our knowledge, this is the first set of results that allows any type of random forest, including classification and regression forests, to be used for provably valid statistical inference. In experiments, we find causal forests to be substantially more powerful than classical methods based on nearest-neighbor matching, especially in the presence of irrelevant covariates.