Date:         Tue, 2 Sep 2008 10:08:12 -0400
Reply-To:     Kevin Viel <citam.sasl@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Kevin Viel <citam.sasl@GMAIL.COM>
Subject:      Re: Class variables proc logistic

On Tue, 2 Sep 2008 04:46:27 -0700, tal <talila.ar@GMAIL.COM> wrote:

>Hi! I use class variables in proc logistic , but I have the following
>problem: I want to do an explanatory variable that counts the number
>of missing values in var1-var20 (i'll call it num_miss)
>As I use the class statement => this variable num_miss is a linear
>combination of var1-var20, but this is only artificially- because the
>category "missing value" of var1-var20 is not a real explanatory
>variable
>Is there anyway to enter that num_miss variable without having all
>this troubles?

Aside from mechanics, you might receive much more help if you describe your
data.  Tell us what the variables are or, if confidentially is necessary,
what it means to be missing and why you want num_miss.  Also, tell us how
you sampled these data, specifically, whether you employed stratified
sampling.

HTH,

Kevin