Hi:
I suspect you could test this with a structural equation modeling
program (using Maximum Likelihood estimation) like glamm (I use Mplus
for those purposes because it is much faster with ML estimation than
glamm is; however, I don't know glamm well enough to make a definitive
statement on this). Basically, in the case of a structural-equation
modeling program you would define a stacked (multisample or multiple
group model) and the test for the equality of coefficient/2 across
groups (don't forget to allow the disturbances of y x1 and x2 to
correlate, otherwise you will get OLS estimates).
HTH,
John.
____________________________________________________
Prof. John Antonakis, Associate Dean
Faculty of Business and Economics
Department of Organizational Behavior
University of Lausanne
Internef #618
CH-1015 Lausanne-Dorigny
Switzerland
Tel ++41 (0)21 692-3438
Fax ++41 (0)21 692-3305
Faculty page:
http://www.hec.unil.ch/people/jantonakis
Personal page:
http://www.hec.unil.ch/jantonakis
____________________________________________________
On 01.01.2010 23:33, Sungbok Lee wrote:
> Dear all,
>
> Is there anyone that have experienced the same question with mine in
STATA?
> Please let me know how you did it.
>
> My question is to test the equality of the two IV estimators. But I
have no
> idea of how to define an unrestricted regression expression in STATA.
>
> *restricted model: all coefficients are equal between the two groups
> ivreg y (x1 x2 = z1 z2)
>
> *unrestircted model: the coefficient for x1 is different but the
coefficient
> for x2 is equal
>
>
> javanfish.
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/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/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
