It looks like this is a convergence issue though it is difficult to
determine what the problem is without knowing the exact syntax and
other information (traceback() etc.)
Kosuke
---------------------------------------------------------
Kosuke Imai Office: Corwin Hall 041
Assistant Professor Phone: 609-258-6601
Department of Politics Fax: 973-556-1929
Princeton University Email: kimai(a)Princeton.Edu
Princeton, NJ 08544-1012
---------------------------------------------------------
On Aug 19, 2005, at 3:49 PM, Gary King wrote:
On Fri, 19 Aug 2005, Igor Grigorovich wrote:
Dear Dr. King,
I was excited to find your package Zelig for Multinomial Logistic
Regression for Dependent Variables with Unordered Categorical
Values, which I used to model ecoregions as a function of
geographic coordinates. Unfortunately, I was not able to complete
my calculations due to the following error:
Error in tfun(mu = mu, y = y, w = w, res = FALSE, eta = eta, extra) :
NAs are not allowed in subscripted assignments
In addition: Warning messages:
1: there are NAs here 3423 in: slot(family, "inverse")(eta, extra)
2: fitted values close to 0 or 1 in: tfun(mu = mu, y = y, w = w,
res = FALSE, eta = eta, extra)
When using help, I was suggested to check the option 'FOO(PKG)
TITLE' in the (FOO, package = PKG), however I was not able to do
so. Although R is a new environment for me, however I was able to
use successfully other R package (mgcv). Perhaps this is not a
software problem; my model possible cannot be converged.
I would appreciate any your suggestions. Hopefully this feedback
is of help,
Yours sincerely,
Igor Grigorovich
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options:
http://lists.hmdc.harvard.edu/?info=zelig
Zelig program information:
http://gking.harvard.edu/zelig/
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options: