Date: Tue, 25 Oct 2005 11:18:23 -0700
Reply-To: Daniel <daniel.biostatistics@GMAIL.COM>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Daniel <daniel.biostatistics@GMAIL.COM>
Subject: Multinomial Confidence Intervals
Content-Type: text/plain; charset="iso-8859-1"
Hello to All,
I would like to know if there is any *automatic* way to produce
multinomial confidence intervals in SAS. If, for instance, I have a
question which offers 4 different possible choices of answer, I would
like to compute the proportion of subjects who chose each option, as
well as a CI associated with that proportion.
The FREQ procedure will output the desired CI when the BINOMIAL option
is specified in the TABLES statement, but this is only done for 1 level
of the multinomial variable, not all of them.
If no such option exists, Leslie Daly posted a macro to obtain binomial
confidence intervals in 1995 (link below) and I was wondering if this
could be used. Here is the link to the original post: