I would start with a smaller model and see where the model breaks down. My
guess is that this is a problem of your explanatory variables - perhaps
there is a very high collinearity or the model is being saturated in one
way or another.
Kosuke
-----------------------------------------------------
Kosuke Imai Office: Corwin Hall 041
Assistant Professor Phone: 609-258-6601
Department of Politics eFax: 973-556-1929
Princeton University Email: kimai(a)Princeton.Edu
Princeton, NJ 08544-1012
http://imai.princeton.edu
-----------------------------------------------------
On Fri, 20 Apr 2007, Olivia Lau wrote:
---------- Forwarded message ----------
From: Brandon Van Dyck <brandonvandyck(a)gmail.com>
Date: Apr 20, 2007 11:03 PM
Subject: bug in mlogit?
To: olau(a)fas.harvard.edu
Hi. I'm using the latest version of Zelig and VGAM and encountered a
strange error with the mlogit model and interaction terms:
library(Zelig)
# Read in the data
data <- read.csv("ferdi.csv")
z.out2 <- zelig(as.factor(outcome.2) ~ bureau + democracy+
I(democracy^2) + army.1000 +
duration + I(duration^2) + exports + gini + border + ethnicity + wartype
+ UN + forest + mountain + africa + lnpop + logincome, model="mlogit",
data = data)
# the error i get is:
Error in devmu[small] = sy * log(smu) : NAs are not allowed in
subscripted assignments
Can anyone help me out? The data is attached.
Many thanks,
Brandon
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options:
http://lists.gking.harvard.edu/?info=zelig
Zelig program information:
http://gking.harvard.edu/zelig/