Peter,
This isn't how 2SLS aka IV works. If you were doing 2SLS "by hand",
you have to use ALL your exogenous variables in ALL the 1st stage
regressions.
There's a Stata FAQ on precisely this:
http://www.stata.com/support/faqs/stat/ivreg.html
Hope this helps.
--Mark
Date sent: Thu, 31 Oct 2002 12:49:40 +0100
From: "Peter Haan" <[email protected]>
To: [email protected]
Subject: st: ivreg
Send reply to: [email protected]
> Hi,
> I have a question concerning 2SL2 estimation with several instrumented variables.
> my idea is to do the following
> ivreg y k (x= d f) (z= r g), instrumenting x with d and f and z with r and g
> however as far as I understand stata, the program suggests to do the following:
>
> ivreg y k (x z= d f r g), meaning that I have to instrument x and z with the same variables d f r g.
>
> An other way could be to estimate the predicted x and z and then to use the predicted values in the second step.
> Then I had to correct for the standar errors.
>
> My problem is that in fact I want to instrument not only 2 varaibles( x and z), but 24 variables.
>
> thanks fpr you advise
> Peter
>
>
> *
> * 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/
Prof. Mark E. Schaffer
Director
Centre for Economic Reform and Transformation
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS UK
44-131-451-3494 direct
44-131-451-3008 fax
44-131-451-3485 CERT administrator
http://www.som.hw.ac.uk/cert
*
* 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/
