How to estimate a two stage fixed-effects regression

[batanaich]

I have been trying to use the two stage least squares specification to estimate the gravity model of trade using fixed effects model . to be specific, here are my problems, 1. can I generate a new variable using the cross-section individual effects and use it in the two stage estimation. 2. can I have a step by step procedure to carry out this process. I have been stuck for almost a fortnight now

thanks

[startz]

If y and x are your LHS and RHS variables, z is your instrument, and q holds the cross-section identifiers, you might try

Code: Select all

tsls y x @expand(q) @ z @expand(q)