Dear colleagues
I am having a bit of a problem. I am trying to conduct logistic
regression on a dataset with 32 records - 16 in each outcome group - and 3
covariates. I am aware that this group is fairly small and, in fact, I
did not get not even a hint of significance.
I attempted exact and Montecarlo logistic regression using LogXact but the
software is giving me only asymptotic estimates (which corresponds with
the result of the conventional logistic regression I did in Stata). The
software refuses to generate exact or Montecarlo estimates.
Unfortunately, my SAS is very rusty and, currently, I do not have access
to the software, so I can't use the SAS routine. I have to admit I am
more of a STATA user.
Can you help me identify what's wrong? Any suggestions?
Thanks
Neville Calleja
|