To my modest knowledge about this topic, ivreg commands are used for
instrumental variable models rather than treatment effects model as
treatreg or biprobit do.
I am using a treatment effects model.
Thanks,
Luca
-----Messaggio originale-----
Da: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] Per conto di Austin
Nichols
Inviato: 31 August 2007 15:59
A: statalist@hsphsun2.harvard.edu
Oggetto: Re: st: info on treatment effects model
Luca--
You can use -ivreg2- or -ivregress-; the fact that the endogenous
variable/s is/are binary does not affect consistency of the IV
estimator, and you can generate new instruments to improve efficiency
(but beware of pernicious weak instruments problems). See e.g. pages
49-50 of
http://www.stata.com/meeting/6nasug/causal.pdf
On 8/31/07, Luca Tiberti <luca.tiberti@unifi.it> wrote:
> I have a model where the dependent variable is continuous and more
than one
> regressor endogenous and binary. I know that with only one endogenous
> regressor I should use the command "treatreg"; which command is
suitable
> with more than one endogenous binary regressor?
*
* 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/
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.484 / Virus Database: 269.13.0/980 - Release Date:
30/08/2007 18:05
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.5.484 / Virus Database: 269.13.0/980 - Release Date:
30/08/2007 18:05
*
* 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/