Marginal effect and its CI using ivreghdfe

[Jorge Pino Lagos]

Hi,

I am estimating a two-way fixed effect IV regression with an interactive variable as a predictor. I would like to obtain the marginal effects (which is variable depending on the level of the interaction term) but there is an incompatibility with the command.

I have been looking for another way to get the marginal effects and the confidence intervals but I have not had luck.

Is there some other way to get the ME and its CI? I'm currently doing by hand but it is not easy the algebra.

Thank you in advance,
Jorge.

[Andrew Musau]

ivreghdfe is from SSC, as you are asked to explain (FAQ Advice #12). margins is compatible with the command, the only thing is that you will get a warning error. If you know what you are doing, specify the -force- option to override this error.