[Limdep Nlogit List] (no subject)

[William Greene]

Aliza. LIMDEP does not have a routine for such a model. Note that there is
no way to solve for a reduced form the way you have stated it.  Indeed, it
is unclear what the actual equations would look like. I.e., how do Y and W
appear in the density for the NB equation?  This is not regression. I can't
say I see a way to fit such a model. If any of the 3 equations were in reduced
form, there might be a starting point using some kind of prediction (though
not for the Y equation which has no natural prediction).  But, without that,
I'm open to suggestions.
Regards,
Bill Greene

----- Original Message -----
From: "Aliza Fleischer" <fleische at agri.huji.ac.il>
To: limdep at limdep.itls.usyd.edu.au
Sent: Tuesday, January 8, 2008 3:06:30 AM (GMT-0500) America/New_York
Subject: [Limdep Nlogit List] (no subject)

 
We wish to estimate the following simultaneous equations model:
 
w=w(y,q,x1)
q=q(y,w,x2)
y=y(w,q,x3)
 
where the 'w' equation is tobit, the 'q' equation is negative binomial and
the 'y' equation is ordered probit.
 
Does Limdep have a built-in routine for such a model?
 
Any help is greatly appreciated.
 
Best regards,
 Aliza 
 
Aliza Fleischer
Department of Agricultural Economics and Management
Hebrew University of Jerusalem
P.O.Box 12, Rehovot 76100
Israel
Tel: +972-8-9489144
Fax: +972-8-9466267
http://agri3.huji.ac.il/~fleische/
 
 
_______________________________________________
Limdep site list
Limdep at limdep.itls.usyd.edu.au
http://limdep.itls.usyd.edu.au


-- 
Professor William Greene
Department of Economics
Stern School of Business
New York University
44 West 4th St., Rm. 7-78
New York, NY   10012
http://www.stern.nyu.edu/~wgreene
212.998.0876