Date:         Mon, 5 Mar 2012 13:43:27 -0500
Reply-To:     Rick Wicklin <Rick.Wicklin@SAS.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Rick Wicklin <Rick.Wicklin@SAS.COM>
Subject:      Re: output from regression?

On Mon, 5 Mar 2012 15:09:56 +0000, Zdeb, Michael S <mzdeb@ALBANY.EDU>
wrote:

>hi ... I think that the original posting wanted to know if there was some
way to create a data set of observations that were used by the PROC in the
MODEL creation rather than just produce summary stats
>
>data nomiss;
>set mydata;
>if ^cmiss(of _all_);
>run;
>

I'm sure Mike is aware of this, but when you use this approach you'll want
to avoid the _ALL_ or _NUMERIC_ keywords. If you use ^CMISS(of _ALL_) it
will exclude observations for which the response variable is missing. The
regression procedures only exclude observations for which the explanatory
vars are missing (missing response is OK), so you'll have to explicitly
name the explanatory variables:
   if ^cmiss(of x1-x10);
or
   if ^cmiss(x1, x7, x10);

Rick Wicklin