just a guess, but it sounds like you have fewer observations than needed
for the number of variables you have. p*(p+3)/2 is the number of
parameters, with p=num of vars. that needs to be << n.
Gary
On Tue, 13 Nov 2007, ian.a.robinson(a)studentmail.newcastle.edu.au wrote:
Hi there,
I am new to MI and was wondering if I could obtain some help.
At the moment, I am performing MI on a dataset of size n= 200. This appears to work
fine. However, part of my project is to break up the data set into male (n=92) and female
(n=108). When I try and run the same MI for the male and female data sets using the
Ameila package, I get the following errors
For the male data set, the error message:
"The resulting variance matrix was not invertible. Please check your data for
highly collinear variables.
Warning message:
In amelia.prep(data = data, m = m, idvars = idvars, empri = empri, :
You have a small number of observations, relative to the number, of variables in the
imputation model. Consider removing some variables,"
and for the female data set:
"The resulting variance matrix was not invertible. Please check your data for
highly collinear variables."
As a result, if I want to perform m = 50 imputations, R will only provide me with 40 or
so.
Have others had similar problems? If so how have you overcome them. If not, what can
you suggest I do?
Regards,
Ian
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=amelia
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: