Date:         Wed, 14 Nov 2007 11:43:00 -0600
Reply-To:     "data _null_," <datanull@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "data _null_," <datanull@GMAIL.COM>
Subject:      Re: transform data
Comments: To: Jerry <greenmt@gmail.com>
In-Reply-To:  <200711141633.lAEERcSY013907@mailgw.cc.uga.edu>
Content-Type: text/plain; charset=ISO-8859-1

On Nov 14, 2007 10:33 AM, Jerry <greenmt@gmail.com> wrote:
> There are only 3 different values for the variable "code", but a lot of
> different values (in thousands) for the variable "ID".

I think (in thousands) is important and may have been overlooked.

This example performs well with IDs close to 1M.


1807  data work.one;
1808     do id=1 to 1e6;
1809        do code = 'a','b','c';
1810           do n = 1 to rantbl(12345,.1,.5,.25,.1)-1;
1811              output;
1812              end;
1813           end;
1814        end;
1815     run;

NOTE: The data set WORK.ONE has 4501860 observations and 3 variables.
NOTE: DATA statement used (Total process time):
      real time           2.26 seconds
      cpu time            2.07 seconds


1816  data work.counts;
1817     do until(last.id);
1818        set work.one;
1819        by id;
1820        array _c[3] code_a code_b code_c;
1821        _c[indexC('abc',code)] + 1;
1822        end;
1823     output;
1824     call missing(of _c[*]);
1825     run;

NOTE: There were 4501860 observations read from the data set WORK.ONE.
NOTE: The data set WORK.COUNTS has 998963 observations and 6 variables.
NOTE: DATA statement used (Total process time):
      real time           2.46 seconds
      cpu time            2.46 seconds