Date:         Thu, 30 Jul 2009 14:02:34 -0400
Reply-To:     Gerhard Hellriegel <gerhard.hellriegel@T-ONLINE.DE>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Gerhard Hellriegel <gerhard.hellriegel@T-ONLINE.DE>
Subject:      Re: Create new data sets
Content-Type: text/plain; charset=ISO-8859-1

if you know how the rank is calculated, do so and store the rank in a new
variable r. Then sort it by r;
Gerhard


On Thu, 30 Jul 2009 10:52:00 -0700, oslo <hokut1@YAHOO.COM> wrote:

>Hi all,

Suppose that I have a data set named "olddata" and I would like to create
a new data set from olddata by changing rank of data but without chancing
the correlation and regression between variables. Example

data oldata;
input�x1 x2 x3 y;
cards;
2 3 4 5�
3 4 4 6
5 7 3 7
2 6 1 5
1 4 6 7
1 2 3 9

Now I would like get new data set from the olddata�only changing the place
of rows. That is the third row in olddata could be the 5th row of new
data, likewise th first of the olddata could be 4th row of the new data.

I would be greatly appriciated geting your helps,

TIA
Oslo