re:st: Re: Quadratic Instrumental Variables

[Mojgan Stegl <mojgan.stegl@uni-tuebingen.de>]

Thank you Christoper for your helpful hints you have kindly referred me to!

Best wishes
Mojgan


Zitat von Christopher Baum <kit.baum@bc.edu>:

> <>
> Is it possible to use quadratic instrumental variables on Stata using
>   ivregress command?
>
> fertility = a0 +a1 x education + a2 x ...+ u
>
> education is endogenous so i will use distance to Tehran and square
> distance to Tehran as instruments. The reduced form equation:
>
> education = b0 + b1x distance to Tehran + b2 x (distance to Tehran)^2
>
>
>
>
> This is not 'nonlinear IV' in any sense. If distance is exogenous  
> (as it obviously is) then
> including it and any nonlinear function of it is perfectly  
> straightforward IV. You're just
> saying that education is a curvilinear function of distance to the  
> capital (and whatever
> included exogenous factors appear in the main equation).
>
> To get a broader set of diagnostics for your IV model, you might  
> want to use Baum-Schaffer-Stillman's
> -ivreg2-, latest version on SSC, documented extensively in Stata J  
> 2003, 2007 (free downloads).
>
> Kit
>
>
> Kit Baum   |   Boston College Economics & DIW Berlin   |    
> http://ideas.repec.org/e/pba1.html
>                              An Introduction to Stata Programming  |  
>   http://www.stata-press.com/books/isp.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/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/