Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: how to test equality of coefficients - ivreg e suest?


From   John Antonakis <John.Antonakis@unil.ch>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: how to test equality of coefficients - ivreg e suest?
Date   Thu, 14 Feb 2013 18:02:23 +0100

Ah....OK. Then set it up as a multiple group SEM (i.e., you have two groups/samples of observations and you want to check for cross group parameter constraints). This is like a -suest-, but you can do multiple equations.

See -help sem-

Best,
J.

__________________________________________

John Antonakis
Professor of Organizational Behavior
Director, Ph.D. Program in Management

Faculty of Business and Economics
University of Lausanne
Internef #618
CH-1015 Lausanne-Dorigny
Switzerland
Tel ++41 (0)21 692-3438
Fax ++41 (0)21 692-3305
http://www.hec.unil.ch/people/jantonakis

Associate Editor
The Leadership Quarterly
__________________________________________

On 14.02.2013 17:16, Manuela Deidda wrote:
Dear John,
thank you for your quick  reply.
However, I have some doubts regarding this solution.
  Indeed, the gmm solution in the thread works when, given the
dependent variable Y and the indicator variable I (determining whether
Y belongs to group 1 or to group 2):

Y=Y1 if I=1 and 0 otherwise
Y=Y2 if I=0 and 0 otherwise

My situation is more like this:
Y=Y1 if I=1 and missing otherwise
Y=Y2 if I=0 and missing otherwise

In the first case, one estimates coefficients through the whole sample
(simply, you have that the variable is truncated at zero).
In the second case, you want only to estimate coefficients in the
subsample where you estimate the variable.

Ex. if n=1000
I=1 if _n=1/250
I=0 if _n=251/1000

I simply need to estimate Y1 when I have 250 observations and Y2 when
I have the other 750 obs...

How can I deal with this?

Best,
M.


2013/2/14 John Antonakis <John.Antonakis@unil.ch>:
Try

http://www.stata.com/statalist/archive/2011-11/msg00487.html

This is a very elegant and quick solution:

Best,
J.

__________________________________________

John Antonakis
Professor of Organizational Behavior
Director, Ph.D. Program in Management

Faculty of Business and Economics
University of Lausanne
Internef #618
CH-1015 Lausanne-Dorigny
Switzerland
Tel ++41 (0)21 692-3438
Fax ++41 (0)21 692-3305
http://www.hec.unil.ch/people/jantonakis

Associate Editor
The Leadership Quarterly
__________________________________________


On 14.02.2013 13:02, Manuela Deidda wrote:
is there any way to compare equality of estimated coefficients after
ivreg?
I have to estimate an iv regression in two different subsamples, defined
by
the dummy variable I:
1) I=1
2) I=0

   with regress one can use suest to compare coefficients in a sample
split.
Is there a way to do something similar after ivreg?

Any help is appreciated,
thanks a lot!
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/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/faqs/resources/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/faqs/resources/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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index