I have following model:
eprobit y x, entreat(trtvar = x z)
I suspect my treatment variable is endogenous so I use z as instrument to solve this problem.
Do this also solve the problem of Stable Unit Treatment Value Assumption (SUTVA) ?
I think SUTVA is crucial assumption in treatment effect model with Conditional Independence assumption (CIA).
eprobit y x, entreat(trtvar = x z)
I suspect my treatment variable is endogenous so I use z as instrument to solve this problem.
Do this also solve the problem of Stable Unit Treatment Value Assumption (SUTVA) ?
I think SUTVA is crucial assumption in treatment effect model with Conditional Independence assumption (CIA).
