As for causal treatment effect estimation, Difference-in-difference is not for heterogenous individual treatment and Synthetic control method is not a parametric model. I am thinking to design Tree-based Difference-in-Difference algorithm.
The paper “Residual Weighted Learning for Estimating Individualized Treatment Rules” (Journal of the American Statistical Association, 2017), seems very potential. But I did not figure out implementation.
Can anyone give me suggestions to estimate heterogenous individual treatment effect?