Date:         Wed, 21 Jul 2010 12:46:06 -0700
Reply-To:     Dale McLerran <stringplayer_2@YAHOO.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Dale McLerran <stringplayer_2@YAHOO.COM>
Subject:      Re: summary statistics of all variables
In-Reply-To:  <201007211902.o6LGcEfr013674@willow.cc.uga.edu>
Content-Type: text/plain; charset=us-ascii

The presentation that PROC MEANS directs to the output window
can be generated to an ascii file by using PROC PRINTTO.  It is
then a matter of reading the ascii file and parsing the records
to determine where the actual data begin and end.  I submitted
the following code and obtained the desired results.


/* Route output to an ascii file */
proc printto print="c:\temp\sumstats.lst";
run;

/* Request statistics employing PROC MEANS */
options formchar='-';
proc means data=test2 nmiss n min max;
  var _numeric_;
run;

/* Reset output to the default location */
proc printto;
run;

/* Parse the ascii file employing the following steps */
data sumstats;
  length vname $ 32 line $ 200;
  infile "c:\temp\sumstats.lst" length=linelen end=lastrec;
  input _tmp $char1. @;
  input @1 line $varying200. linelen;
  if left(line)=:"-----" then do until(lastrec);
    input _tmp $char1. @;
    input @1 line2 $varying200. linelen @;
    if left(line2)^=:"-----" then do;
      input @1 vname nmiss n min max;
      output;
    end;
  end;
  keep vname nmiss n min max;
run;

/* Print the data set which contains the summary statistics */
proc print data=sumstats;
run;


That we cannot get the statistics in a vertical file directly
is a real pain in the patootie.

Dale

---------------------------------------
Dale McLerran
Fred Hutchinson Cancer Research Center
mailto: dmclerra@NO_SPAMfhcrc.org
Ph:  (206) 667-2926
Fax: (206) 667-5977
---------------------------------------


--- On Wed, 7/21/10, Ya Huang <ya.huang@AMYLIN.COM> wrote:

> From: Ya Huang <ya.huang@AMYLIN.COM>
> Subject: Re: summary statistics of all variables
> To: SAS-L@LISTSERV.UGA.EDU
> Date: Wednesday, July 21, 2010, 12:02 PM
> Mike,
>
> "The only issue is that the data set is as linear as any
> made-for-TV movie."
> that is exactly why _null_, another Mike and me are trying
> to avoid.
> We all trying to work out a solution that a vertical
> dataset can be
> created without doing many other steps, such as transpose
> etc.
>
> Ya