Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Levels of instruments in fixed effects xtivreg


From   "Mark Schaffer" <[email protected]>
To   [email protected]
Subject   Re: st: Levels of instruments in fixed effects xtivreg
Date   Thu, 09 Sep 2004 11:19:16 +0100

Katie,

Date sent:      	Wed, 08 Sep 2004 13:13:49 -0400
From:           	Katie Winder <[email protected]>
Subject:        	st: Levels of instruments in fixed effects xtivreg
To:             	Statalist <[email protected]>
Send reply to:  	[email protected]

> Dear Statalist,
> I am estimating a wage equation with several endogenous regressors using
> xtivreg, fe using instruments Z.  I would like to include levels of Z as
> instruments as well as mean deviations of Z, but xtivreg, fe seems to
> automatically difference the instruments along with the other variables.
> Does anyone know of a way to trick Stata into including levels as well as
> differences?

I'm not sure I understand, or if this makes sense with fixed effects.

(a) xtivreg,fe is equivalent to (b) standard IV with fixed effects 
added by hand, and (c) demeaning all your variables (dependent, 
regressors, excluded instruments) and then running standard IV.

You can see this from the following silly example using the auto 
dataset.  The variables prefixed with c_ are in mean-deviation form 
(created using Ben Jann's -center- command).  -foreign- is the fixed 
effect.

sysuse auto
by foreign: center price mpg weight, prefix(c_)
xtivreg price (mpg=weight), fe i(foreign)
ivreg2 price foreign (mpg=weight)
ivreg2 c_price (c_mpg=c_weight)

The coefficient on mpg is the same in all 3 estimations.

The problem is that I can't see what it means to increase the 
instrument list in the way you suggest, because you get 
collinearities and the extra instruments drop out.  Try it with

ivreg2 c_price (c_mpg=c_weight weight)

and you'll see that you get exactly the same results as before.

Cheers,
Mark

> 
> Thanks!
> 
> Regards,
> Katie Winder
> Johns Hopkins U.
> Dept. of Economics
> 
> *
> *   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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index