Reply to commentary by Imai, Keele, Tingley, and Yamamoto concerning causal mediation analysis.
Reply to commentary by Imai, Keele, Tingley, and Yamamoto concerning causal mediation analysis.
复制标题
回复 Imai、Keele、Tingley 和 Yamamoto 关于因果中介分析的评论。
DOI：
10.1037/met0000022
复制
发表时间：
2014
影响因子：
7
通讯作者：
Pearl,Judea
中科院分区：
文献类型：
--
作者：
Pearl,Judea
这一评论阐明了结构因果模型如何统一调解的图形和潜在结果方法，以及为什么所产生的调解公式在两个框架中是相同的。它进一步解释了在什么情况下基于可知性的假设是过度限制的，以及为什么在判断这些假设是否可信之前，这些假设需要图形解释。最后，该评论解释了传统和现代因果调解方法之间的关键区别，并说明了为什么调解的概念需要正确定义反事实条件句，而不是贝叶斯条件句。
This comment clarifies how structural causal models unify the graphical and potential outcome approaches to mediation, and why the resulting mediation formulas are identical in both frameworks. It further explains under what conditions ignorability-based assumptions are over-restrictive and why such assumptions require graphical interpretations before they can be judged for plausibility. Finally, the comment explains the key difference between traditional and modern methods of causal mediation, and demonstrates why the notion of mediation requires counterfactual rather than Bayes conditionals to be properly defined.(PsycINFO Database Record (c) 2016 APA, all rights reserved)