2SLS - Specification question

[Gunsee]

Hello everyone,

I am currently reviewing an old statistical work where Two-stage least squares has been used. Now, this is an area I am not particular familiar with and in this model they have specified the model as:

Code: Select all

log(wage) = β0 + β1Education + β2exper + β3exper^2 + u

The IV for education is Mother´s education (Moth_edu) and Father´s education (Fath_edu) as well as years of work experience (work_years) and years of work experience squared (work_yrssqr).

The result (printed in Eviews) looks like this:

http://imageshack.us/photo/my-images/690/2slso.jpg/

My question is, is it possible to include work_years and work_yrssqr into the regression given that they are exogenous?


I am trying to get a sense of this as this specification and result feels very counterintuitive so please help me out :)