st: Fwd: instrumental variables and linear constraints

[Marcello Pagano <pagano@hsph.harvard.edu>]

Begin forwarded message:

> From: Kit Baum <baum@bc.edu>
> Date: June 19, 2007 11:32:01 AM EDT
> To: statalist@hsphsun2.harvard.edu
> Subject: re: instrumental variables and linear constraints
>
>
> Florian said
>
> Could someone tell me how I can estimate an IV regression with a  
> linear
> constraint on the coefficients of endogenous variables (since the  
> option
> "constraints" is not allowed with ivreg)?
>
> -ivreg2- does not deal with constraints, either (nor does - 
> regress-; -cnsreg- does). I would suggest that you algebraically  
> substitute the constraints into the equation. If you want to  
> untangle the coefficients thereafter, -lincom- can do so, and  
> provide a std.err. for the missing coeffs.
>
> Example from economics with two-factor Cobb-Douglas (constant  
> elasticity) production fn, with constraint of constant returns to  
> scale (CRTS), implying slope coeffs sum to unity (all variables  
> assumed to be logarithms):
>
> // unconstrained
> regress output labor capital
> // impose constraint b_labor + b_capital = 1
> g ol = output-labor
> g kl = capital-labor
> regress ol kl
> // recover labor coefficient
> lincom 1-kl
>
>
>
>
> Kit Baum, Boston College Economics and DIW Berlin
> http://ideas.repec.org/e/pba1.html
> An Introduction to Modern Econometrics Using Stata:
> http://www.stata-press.com/books/imeus.html
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/