Date:         Fri, 13 Oct 2000 11:30:47 -0600
Reply-To:     Jack Hamilton <JackHamilton@FIRSTHEALTH.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Jack Hamilton <JackHamilton@FIRSTHEALTH.COM>
Subject:      Re: Duplicate Observations!
Comments: To: carlytuur@MY-DEJA.COM
Comments: cc: ylau1@irix2.gl.umbc.edu
Content-Type: text/plain; charset=us-ascii

There are several circumstances under which NODUPLICATES
fails without warning.  Please see my paper, "The Problem
With NODUPLICATES", at

  <http://www2.sas.com/proceedings/sugi25/25/po/25p221.pdf>.

Incidentally, the code below won't run; PROC SORT requires a
BY statement.  Code that would work:

   proc sort data=data1 nodup;
      by _all_;
   run;



--
JackHamilton@FirstHealth.com
Development Manager, Technical Group
METRICS Department, First Health
West Sacramento, California USA


>>> <carlytuur@MY-DEJA.COM> 10/13/2000 5:34 AM >>>
In my opinion the easiest solution is to use a sort;

proc sort data = data1 nodup;
run;

This compares a complete record with the previous record. Removing it
when all variables are the same.

Pascal

In article <8rl1ul$sio$1@news.umbc.edu>,
  <ylau1@irix2.gl.umbc.edu> wrote:
> Hi. Does anyone know how to obtain recrods that have duplicates? For
example,
>
> Obs   Var1    Var2    Var3    Var4    Var5    Var6    Var7
> 1     2869    D368    4103    1988    1988    1999    2063
> 2     2869    D368    4103    1988    1988    1999    2063
> 3     5292    D369    0010    1988    1988    1992    2063
> 4     5292    D369    0003    1988    1988    1992    2063
> 5     9232    6944    1892    1988    1988    1991    2063
> 6     9232    6944    1892    1988    1988    1991    2063
> 7     4304    D008    3005    1991    1991    1999    2063
> 8     4304    D008    3005    1991    1991    1999    2063
> 9     4304    D008    7009    1991    1991    1999    2063
> 10    2096    4108    2800    1988    1992    1998    9063
> 11    4096    7016    2800    1988    1992    1998    2063
>
> I would like to keep observations 1,2,5,6,7, and 8.
>
> Thanks for any help.
>
> Anna
>


Sent via Deja.com http://www.deja.com/
Before you buy.