Fwd: st: 2SLS and eivreg

[Henrik Andersson <henrik.andersson@nek.lu.se>]

Hi,

I would like to apologize for a "typo" in the mail I posted earlier. (See 
below.)

The dependent variable should not be the same as one of the explanatory 
variables, i.e. the equation

y=ax+by+cz+e

should instead had been written as

p=ax+by+cz+e

Best regards

Henrik Andersson


> X-Authentication-Warning: hsphsun2.harvard.edu: majordom set sender to 
> owner-statalist@hsphsun2.harvard.edu using -f
> X-Sender: nekhan@pandora.nek.lu.se
> X-Mailer: QUALCOMM Windows Eudora Version 5.2.1
> Date: Wed, 18 Aug 2004 19:28:21 +0200
> To: statalist@hsphsun2.harvard.edu
> From: Henrik Andersson <henrik.andersson@nek.lu.se>
> Subject: st: 2SLS and eivreg
> Sender: owner-statalist@hsphsun2.harvard.edu
> Reply-To: statalist@hsphsun2.harvard.edu
>
> Hi,
>
> I am estimating the following equation
>
> y=ax+by+cz+e
>
> where a,b and c are coefficients to be estimated, x, y and z my variables, 
> and e the error term. My x is correlated with e and I therefore use 2SLS 
> (ivreg). I also suspect that I have measurement errors in z and would 
> therefore, as a sensitivity test, like to use the regression command 
> eivreg (errors-in-variables regression).
>
> Is it possible to run a combined 2SLS-eivreg, i.e. is it possible to 
> instrument x at the same time as I run eivreg, or vice versa?
>
> I (still) use Stata 7.
>
> Best regards
>
> Henrik Andersson
>
> *
> *   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/
******************************************
Henrik Andersson
Dept. of Economics
Lund University
P.O. Box 7082
SE-220 07 Lund

Tel: +46 46 222 30 72
Fax: +46 46 222 06 51
mailto:henrik.andersson@nek.lu.se 
****************************************** 
*
*   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/