LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (December 1999, week 3)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Mon, 20 Dec 1999 03:34:29 GMT
Reply-To:     SAlbert <salbert@AOL.COMSTOPSPAM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         SAlbert <salbert@AOL.COMSTOPSPAM>
Organization: AOL http://www.aol.com
Subject:      Re: random sampling

Ju-Tsung Pi wrote:> > I have a data set of 887 observations from a normal distribution. I >would like to draw a > sample of 250 observations out ot 887 obs. How can I do this? > Thanks! >

If you only need approximately 250, you can use code like the following:

data two; set one(where=(ranuni(0) < 250.0/887)); run;

If you need an exact 250, you could try several approaches, but the easiest would be to add a random variable, sort on it, and then take the first 250 cases: data two; set one; dummy = ranuni(0); run; proc sort data=two; by dummy; run; data three; set two(obs=250); run;

Note: code above has not been tested.

Steve Albert


Back to: Top of message | Previous page | Main SAS-L page