Dear All:
Just to add to the discussion, I wonder why no one mentions Maddala
(1983)"Limited dependent and qualitative variables in econometrics." In
pages 243-245, Maddala explains what to do when there is a dichotomous,
endogenous variable. Basically, these are the steps:
Eq1: y1 = a1*x + a2*y2 + v1
Eq2: y2 = b2*x + v2, where y2 is binary and endogenous.
Transform y2 into y3 = y2 / square root of sigma2, where sigma2 is the
variance of v2 (i.e., the variance of the residuals from regressing the
endogenous binary variable, y2, on all the exogenous variables in equation 1
using probit ML).
Get the predicted y3.
Then estimate equation 1 above by OLS with the predicted y3 replacing y2.
Could anyone comment on this method? I am not an econometrician, and
therefore need some confirmation on my understanding of Maddala, too.
The one problem I see with Maddala's method is how to interpret the
coefficients of a2 because y2 has been transformed.
Thanks
Hinh
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of
[email protected]
Sent: Friday, June 08, 2007 12:50 PM
To: [email protected]
Subject: RE[2]: st: biprobit with endogenous binary
Dear Partha,
thanks for your reply. Could you kindly be more detailed in your reply
because I did not understand very well. Sorry.
Yesterday I visited
http://www.gseis.ucla.edu/courses/ed231c/notes1/biprobit.html
and in the last example the use the biprobit command instrumenting the
endogenous variable. I do not understand the restrictions you were
speaking about.
Thanks a lot,
Luca
>Yes, subject to the usual parametric assumptions and appropriate exclusion
>restrictions.
>
>Cheers,
>
>Partha
>
>
>Marcello Pagano wrote:
>> For Luca Tiberti:
>>
>>
>>>> Good Day!
>>>> I have a similar system of equations:
>>>>
>>>> y1=x1+x2
>>>> y2=y1+x3+x4
>>>>
>>>> where y1 and y2 are both binary variables and, in particular, y1 is in
>>>> y2 an endogenous regressor and y2 is the main outcome variable. My
>>>> question is whether biprobit is the right command to instrument y1. I
>>>> think yes, but I am not sure.
>>>>
>>>> Thank you very much,
>>>>
>>>> Luca
>>>>
>>
>> *
>> * 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/
>
>--
>Partha Deb
>Department of Economics
>Hunter College
>ph: (212) 772-5435
>fax: (212) 772-5398
>http://urban.hunter.cuny.edu/~deb/
>
>Emancipate yourselves from mental slavery
>None but ourselves can free our minds.
> - Bob Marley
>
>*
>* 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/
