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

[John Antonakis <John.Antonakis@unil.ch>]

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/