Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: iv probit


From   "Brian P. Poi" <bpoi@stata.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: iv probit
Date   Sat, 24 Jun 2006 16:43:13 -0500 (CDT)

On Fri, 23 Jun 2006, Elizabeth Wiemers wrote:

I am under the impression that when an ivprobit command is used in
STATA, the first stage is estimated as a regular OLS, with a continuous
dependent variable, and the second stage is estimated as a probit.  Is
this correct?

Thanks!

Elizabeth,

In the simplest version of the two-step IV probit estimator, yes, the first stage is OLS regression, and the second stage is probit, including both the fitted values from the first stage and the first-stage residuals as regressors (including the residuals is key).

The actual two-step estimator that -ivprobit- uses is a bit more complex, because it is actually a weighted combination of two different estimators; see the Methods and formulas of [R]ivprobit. The first stage, though, is indeed OLS regression.

The maximum likelihood estimator is done in one step; there is no first and second step.

Hope this helps

-- Brian Poi
-- bpoi@stata.com


*
* 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/




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