Date: Wed, 7 Oct 2009 10:53:11 -0400
Reply-To: Peter Flom <email@example.com>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Peter Flom <peterflomconsulting@MINDSPRING.COM>
Subject: Re: predicted probabilities in logistic model
Content-Type: text/plain; charset=UTF-8
ellen <ellenstats@GMAIL.COM> wrote
>I have a logistic regression model containing several co-variates. I now
>wish to estimate the expected number of events for a variable that is not
>in the model but is in my data set. Can I simply add up the predicted
>probabilities for the additional variable and if so how do I obtain the
>confidence interval, can I sum these too??
>Many thanks for any help you can give.
Do you mean an additional observation? I don't see how this could work with additional variables.
If you do mean observations, then have you seen example 51.15 in the SAS/Stat documentation? Title is "scoring data sets with the score statement".
First, run something ala this
ods graphics on;
proc logistic data=Crops plots(only)=effect(x=x3);
model Crop=x1-x4 / link=glogit;
ods graphics off;
proc logistic inmodel=sasuser.CropModel;
score data=Crops out=Score3;
Peter L. Flom, PhD
Website: www DOT peterflomconsulting DOT com