Hi,

I have an unbalanced panel (N=1,350 and T=130), and an outcome variable that is a count (around 6% are zeros). I have a continuous endogenous variable, for which I have a continuous instrument. I know that it's possible to run poisson fixed-effect models (xtpoisson) and poisson models with an instrument (ivpoisson), but I am struggling to find a way to estimate an FE-IV model that combines the two. Any suggestions?

Many thanks!

George

N.B. Similar issues have been raised in the past, without resolution (see https://www.statalist.org/forums/for...-fixed-effects).