ivreghdfe with interactions

[Adrienne Wold]

Hello, I am using the ivreghdfe command as follows:

estimates clear
set more off
eststo : ivreghdfe y (i.x1##c.x2 = i.x1##c.z2), a(year##c.x2)
}

where z2 is my instrument for x2. I get the error "equation not identified; must have at least as many instruments
not in the regression as there are instrumented variables" but have one instrument for x2 so not sure what the problem here is.

[Jeff Wooldridge]

How come you're absorbing year##c.x2? Shouldn't you absorb a unit specific identifier and year? I'm not sure what it means to absorb year##c.x2, but maybe you can tell us more about your application.

[Adrienne Wold]

I'm absorbing year##c.x2 as unit-specific time trends.

[Adrienne Wold]

Creating the interaction terms as a separate variable, things appear to work.