Include same variables used for balancing in xtreg?

[Phil Smith]

Dear experts,

it is my first time running a FE regression and also my first time using entropy balancing for weighting control and treatment group.

I am wondering if I can/have to include the variables I used in the ebalance command in the regression.

KR

[George Ford]

With balancing generally, I've seen it done both ways. The original paper on Coarsened Exact Matching talks a bit about it.

[Cat Santos]

I also have this question. I have also seen it done both ways but I'm still unsure whether it is okay or not to include balancing variables as controls when using entropy balancing.