Date:         Wed, 9 Apr 2003 14:25:05 -0400
Reply-To:     sashole@bellsouth.net
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Paul M. Dorfman" <sashole@BELLSOUTH.NET>
Organization: Sashole of Florida
Subject:      Re: Sampling Question
Comments: To: diskin.dennis@kendle.com
In-Reply-To:  <OF46C1FD04.C1102C55-ON85256D03.005F2ECB@kendle.com>
Content-Type: text/plain; charset="us-ascii"

> -----Original Message-----
> From: diskin.dennis@kendle.com [mailto:diskin.dennis@kendle.com]
>
> Paul,
>
> I'm afraid your algorithim is flawed. It allows duplicates to
> be selected.

Dennis,

The algorithm is perfect; the implementation is plagued with
absent-mindedness: I forgot to surround _n_ with parentheses in

ptr (_p_) = _n_ ;

which should be

ptr (_p_) = ptr (_n_) ;

But wait: Did not I just write the same in response to Dale? I guess I
did. Here goes the memory, again... :-(.

Kind regards,
===================
Paul M. Dorfman
Jacksonville, FL
===================

> From:  Paul Dorfman <paul_dorfman@HOTMAIL.COM>@LISTSERV.UGA.EDU> on
>        04/09/2003 12:36 PM
>
>
> >From: Action Man <wollo_desse@HOTMAIL.COM>
> >
> >I have 7,000 records in my SAS file. Out of these records I want to
> >pick 500 of them randomly, >How do I do that using SAS.
>
> Wollo,
>
> You will no doubt get (or have already gotten) plenty of
> advice how to do it using the "standard" K/N method, where K
> is the sample size and N is the population size. It is based
> on reading all N records from the population file. It is
> plenty sufficient and fast in your case, where N=7000 only
> and K=500 is not a tiny fraction of N. Below is an
> alternative approach allowing to obtain the sample by reading
> K records only, which may be preferable for large Ns (easily
> up to, say, 5E+6 with modern memory sizes) and/or K/N <<
> 1:
>
> %let n_pop  = 7000 ;
> %let n_smpl =  500 ;
>
> data pop ;
>    array vars (*) a b v03-v11 ;
>    do a = 1 to &n_pop ;
>       do b = 3 to 11 ;
>          vars (b) = ceil (ranuni(1) * 1e11) ;
>       end ;
>       output ;
>    end ;
> run ;
>
> data sample (drop = _:) ;
>    array ptr (&n_pop) _temporary_ ;
>
>    do _p_ = 1 to hbound (ptr) ;
>       ptr (_p_) = _p_ ;
>    end ;
>
>    do _n_ = &n_pop to &n_pop - &n_smpl + 1 by -1 ;
>       _p_ = ceil (ranuni(1) * _n_) ;
>       point = ptr (_p_) ;
>       set pop point = point ;
>       output ;
>       ptr (_p_) = _n_ ;
>    end ;
>
>    stop ;
> run ;