[Limdep Nlogit List] Random frontier

William Greene wgreene at stern.nyu.edu
Tue Aug 12 11:44:20 EST 2008


M. Vu. I would guess there are too many random parameters in this model.
The data would generally not be rich enough to support a model with this
many unknowns in it. I would start just with the random first order terms
and buiold it up from there.
/B. Greene

----- Original Message -----
From: "Ha Thu Vu" <hvu at efs.mq.edu.au>
To: limdep at limdep.itls.usyd.edu.au
Sent: Monday, August 11, 2008 8:23:42 PM GMT -05:00 Columbia
Subject: [Limdep Nlogit List] Random frontier

Dear Limdep users,
I am running a random-parameters model for both cost and profit function using the same right hand side variables. It worked with cost function but I could not get through with profit function, the error is: 

Line search does not improve fn. Exit iterations. Status=3
Error   806: Line search does not improve fn. Exit iterations. Status=3

Here are the codes I used: 

FRONTIER;Lhs=LNP;Rhs=ONE, LNY1,LNY2,LNY3,LNW1,LNW2
	,LNY1LNY1,LNY2LNY2,LNY3LNY3, LNW1LNW1,LNW2LNW2
	,LNY1LNY2,LNY1LNY3,LNY2LNY3,LNW1LNW2
	,LNY1LNW1,LNY1LNW2,LNY2LNW1,LNY2LNW2,LNY3LNW1,LNY3LNW2$
FRONTIER;Lhs=LNP ;Rhs=ONE,LNY1,LNY2,LNY3,LNW1,LNW2
	,LNY1LNY1,LNY2LNY2,LNY3LNY3, LNW1LNW1,LNW2LNW2
	,LNY1LNY2,LNY1LNY3,LNY2LNY3,LNW1LNW2
	,LNY1LNW1,LNY1LNW2,LNY2LNW1,LNY2LNW2,LNY3LNW1,LNY3LNW2
	;PDS=7	
	;RPM
	;FCN= ONE(N),LNY1(N), LNY2(N), LNY3(N),LNW1(N), LNW2(N)
	,LNY1LNY1(N),LNY2LNY2(N),LNY3LNY3(N), LNW1LNW1(N),LNW2LNW2(N)
	,LNY1LNY2(N),LNY1LNY3(N),LNY2LNY3(N),LNW1LNW2(N)
	,LNY1LNW1(N),LNY1LNW2(N),LNY2LNW1(N),LNY2LNW2(N),LNY3LNW1(N),LNY3LNW2(N)
	;Eff=PI$

Could you please help as I really got stuck? Thank you very much indeed.
Kind regards,
Ha Thu Vu.


Ha Thu Vu
PhD Candidate in Economics
Department of Economics
Macquarie University, Sydney

_______________________________________________
Limdep site list
Limdep at limdep.itls.usyd.edu.au
http://limdep.itls.usyd.edu.au



More information about the Limdep mailing list