[Limdep Nlogit List] Error: Model is too complex
William Greene
wgreene at stern.nyu.edu
Sat Aug 8 03:23:09 EST 2009
The fact that all alternatives (save I) appear in both
branches is a very bad start. The branches in a nested
logit model cannot overlap. There is a stray comma in
branch AA. Your model is not identified; one of the parameter
pairs in branch AA must be normalized at zero. I would
replaace the ;MODEL:... part of the command with
;RH2=one,X
The tree is something else. I guess you should just
discard branch BB.
/Bill Greene
----- Original Message -----
From: "M. (Muhammad) Sabir" <msabir at feweb.vu.nl>
To: limdep at limdep.itls.usyd.edu.au
Sent: Friday, August 7, 2009 10:16:30 AM GMT -05:00 Columbia
Subject: [Limdep Nlogit List] Error: Model is too complex
Dear all,
I am trying to run the following model ;
Nlogit
; LHS = Choice, Cset, Altij
; Choices = A, B, C, D, E, F, G, H, I
; Tree = AA(A,,B, C, D, E, F, G, H), BB(A, B, C, D, E, F, G, H), CC(I)
; Start = Logit
; Show
; Model :
U(A) = B10+B11*X/
U(B) = B20+B21*X/
U(C) = B30+B31*X/
U(D) = B40+B41*X/
U(E) = B50+B51*X/
U(F) = B60+B61*X/
U(G) = B70+B71*X/
U(H) = B80+B41*X/
U(I) = B90 [0]
;Ivset : (CC) = [1.0]
;Maxit=150
;Utility = U1 $
(All X are individual specific variables and are constant over
different choice).
I get the following error message:
Error 1020: The model is too complex - too many parts. Table overflow.
I could not find this error message in LIMDEP Reference Guide 9.0.
Can someone explain what does this error means or what is wrong here ?
Many thanks.
Kind regards,
Muhammad
_______________________________________________
Limdep site list
Limdep at limdep.itls.usyd.edu.au
http://limdep.itls.usyd.edu.au
More information about the Limdep
mailing list