Date:         Wed, 5 Aug 2009 11:14:42 -0500
Reply-To:     "Data _null_;" <iebupdte@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Data _null_;" <iebupdte@GMAIL.COM>
Subject:      Re: Adding class/dummy variable
Comments: To: Billy Rubin <billy_rubin1618@yahoo.com>
In-Reply-To:  <862981.85730.qm@web44703.mail.sp1.yahoo.com>
Content-Type: text/plain; charset=ISO-8859-1

proc summary nway;
   class x;
   output out=levels(drop=_freq_ _type_ index=(x)) / levels;
   run;
data a;
   set a;
   set levels key=x/unique;
   run;


Or make and INFORMAT etc.

On 8/5/09, Billy Rubin <billy_rubin1618@yahoo.com> wrote:
> My brain is failing me...I know I've done this many times, but I cannot access the knowledge this morning.
>
> How does one create a classification/dummy variable without going through so many data/sql steps.  I have a data table with many string classes and I want to add a numerical classification variable.  What is the trick?
>
>
>
> data a (drop=i);
>  length x $ 2 ;
>  do i = 1 to 50;
>      X = byte(96 + ceil(ranuni(1)*5) ) ;
>          output;
>  end;
> run;
>
> proc sql;
>  create table b as
>  select distinct x
>  from a
>  order by x;
> quit;
>
> data c; set b;
>  CLASS = _n_;
> run;
>
> proc sql;
>  create table d as
>  select a.*, c.*
>  from a, c
>  where a.x = c.x
>  order by x ;
> quit;
>