[Limdep Nlogit List] simultaneous equation
William Greene
wgreene at stern.nyu.edu
Tue May 2 19:39:39 EST 2006
Dear Joe: The program won't know to compute the fitted value as a product
just by the name of the variable. You will have to create it separately with its
own create command. The problem, of course, is what the reduced form is.
If you are confident that the list of exogenous variables is complete, then to do
this using the equivalent of 2SLS, you will have to have separate commands
for each step. Regress Y2 on the X'x, compute the fitted values, compute the
interaction, do the IV regression. (By the way, the algebra of 2SLS doesn't
work here. You can't just plug the fitted interaction term back into the regression,
as you do when you do with linear systems. You have to use IV.)
/Bill Greene
************************************************
Professor William Greene
Department of Economics
Stern School of Business
New York University
44 West 4th St., Rm. 7-78
New York, NY 10012
Ph. 212.998.0876
Fax. 212.995.4218
URL. http://www.stern.nyu.edu/~wgreene
Email. wgreene at stern.nyu.edu
************************************************
----- Original Message -----
From: ZOU Hong <hzou at ln.edu.hk>
Date: Tuesday, May 2, 2006 5:03 am
Subject: [Limdep Nlogit List] simultaneous equation
> Dear All,
>
> I am estimating a system of simultaneous equation consisting of
> three endogenous variables (Y1, Y2, and Y3), my setup is as
> follows:
>
>
>
> SAMPLE; ALL$
>
> CREATE; _YX=Y2*x11 $ an interaction between Y2 and exogenous x11
>
> NAMELIST; E_Y1 =one, Y2, Y3,_YX, x11,x12,
>
> ; E_Y2 =one, Y1, Y3, x21,x22,
>
> ; E_Y3 =one,Y1,Y2,x31, x32,
>
> ; EXOG =x11,x12,
,x21,x22,
,x31,x32,
$
>
>
>
>
>
> 2SLS; Lhs =Y1; Rhs =E_Y1; Inst = EXOG $
>
> 2SLS; Lhs =Y2; Rhs =E_Y2; Inst = EXOG $
>
> 2SLS; Lhs =Y3; Rhs =E_Y3; Inst = EXOG $
>
> STOP
>
>
>
> In Equation Y1(i.e., E_Y1), an explanatory variable is Y2 * X11.
> If I set up the program as above, can LIMDEP correctly handle the
> interaction term (i.e., using the predicted Y2 in constructing the
> interaction term) in 2SLS? If not, how should I set up the program?
>
>
>
> Many suggestions are appreciated
>
> Joe
> _______________________________________________
> Limdep site list
> Limdep at limdep.itls.usyd.edu.au
> http://limdep.itls.usyd.edu.au
>
More information about the Limdep
mailing list