Florian,
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of
> Rodrigo A. Alfaro
> Sent: Tuesday, June 19, 2007 4:28 PM
> To: [email protected]
> Subject: st: Re: Instrumental variables and linear constraints
>
> ///
> What about using "Brute force"? In other words, imposing constraints.
> Maybe using -ivreg2, liml- you can pass constraint to -ml-.
Rodrigo's suggestion could work, but you should use the -cue- rather
than the -liml- option. -ivreg2- doesn't use -ml- for estimating LIML
(the usual way LIML coefficients are found is from the solution to an
eigenvalue problem, and that's what -ivreg2- does). CUE
(continuously-updated GMM) is a generalization of LIML, and does call
-ml-, so you can pass constraints in the way Rodrigo suggests.
Cheers,
Mark
>
> Rodrigo.
>
>
> ----- Original Message -----
> From: "Florian MAYNERIS" <[email protected]>
> To: <[email protected]>
> Sent: Tuesday, June 19, 2007 10:26 AM
> Subject: st: Instrumental variables and linear constraints
>
>
> > Hello,
> >
> > Could someone tell me how I can estimate an IV regression
> with a linear
> > constraint on the coefficients of endogenous variables
> (since the option
> > "constraints" is not allowed with ivreg)?
> >
> > Thank you,
> >
> > Florian Mayneris
> >
> > *
> > * 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/
> *
> * 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/
>
*
* 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/
