ENH: link between treatment effects and Fixed Effects estimators

[josef-pkt]

Kosuke Imai, In Song Kim "On the Use of Linear Fixed Effects Regression Estimators for Causal Inference" http://imai.princeton.edu/research/FEmatch.html
with associated R package https://cran.r-project.org/web/packages/wfe/index.html

shows algebraic equivalence between some matching, IPW, repeated measures treatment effects estimators and fixed effects estimators with binary treatment.

I haven't read it yet.
This should be interesting when we get treatment effects and Panel/FE merged.

(This might also be useful for the link of FE estimators and repeated measures, within ANOVA.)