Direct and Indirect Effects based on Changes-in-Changes
提出一种基于变化中变化假设的新因果中介分析方法，将二元处理对连续结果的效应分解为通过二元中介变量的间接效应和直接效应，并识别了不同子组的平均和分位数效应。
We propose a novel approach for causal mediation analysis based on\nchanges-in-changes assumptions restricting unobserved heterogeneity over time.\nThis allows disentangling the causal effect of a binary treatment on a\ncontinuous outcome into an indirect effect operating through a binary\nintermediate variable (called mediator) and a direct effect running via other\ncausal mechanisms. We identify average and quantile direct and indirect effects\nfor various subgroups under the condition that the outcome is monotonic in the\nunobserved heterogeneity and that the distribution of the latter does not\nchange over time conditional on the treatment and the mediator. We also provide\na simulation study and an empirical application to the Jobs II programme.\n