On Fri, 21 Apr 2000 [EMAIL PROTECTED] wrote:
> I've got a random sample from a population in which a very small
> percentage of elements show a particular characteristic. Those are the
> ones I'm interested in for further study.
>
> My approach is to use a logistic regression on that sample to get an
> estimated probability of having this characteristic. My question is:
> based on that predicted probability, how do I divide my sample into
> those elements that are likely to have that characteristic from those
> that are not (can I determine some cutoff in the predicted
> probability)?
Do you really mean, "How do I divide my sample into those elements that
are likely ...", or are you interested in sampling again from the
origianl population, so as to increase the likelihood of that
characteristic? If the _sample_, you can select all the cases that have
that characteristic, and pursue your inquiries with them.
If your interest lies in generalizing to the population from which your
data were sampled, you presumably have predictors (you haven't mentioned
them, but you can't have been doing logistic regression without 'em):
use them to identify subgroups from which to sample.
More than that it is not possible to suggest with the paucity of
information supplied.
-- DFB.
------------------------------------------------------------------------
Donald F. Burrill [EMAIL PROTECTED]
348 Hyde Hall, Plymouth State College, [EMAIL PROTECTED]
MSC #29, Plymouth, NH 03264 603-535-2597
184 Nashua Road, Bedford, NH 03110 603-471-7128
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
