Date:         Fri, 5 Nov 2010 17:11:32 -0400
Reply-To:     Nat Wooding <nathani@VERIZON.NET>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Nat Wooding <nathani@VERIZON.NET>
Subject:      Re: using data from two data sets
In-Reply-To:  <001601cb7d29$4810e8f0$d832bad0$@com>
Content-Type: text/plain; charset="US-ASCII"

Well, if memory serves, you are calculating the corrected sums of squares so
why not let Proc Means do it all at once:

Data start;
input x ;
z= (x - 2.5)**2;
y + z;
cards;
1
2
3
4
;
Proc Means data = start;
   output out = end  css=css  ;
var x;
   run;


I threw in the y and z variables so that I could check the result.

Nat Wooding

-----Original Message-----
From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of
bbser2009
Sent: Friday, November 05, 2010 4:38 PM
To: SAS-L@LISTSERV.UGA.EDU
Subject: using data from two data sets

For illustration, consider two data sets,



temp1



x

1

2

3

4



temp2 (the mean of x in temp1)

mean

2.5



Now I would like to sum all the squares like this:

(1-2.5)**2+(2-2.5)**2+...



Apparently we need data from the two different data sets, but I do not want
to combine the two into one.

I just need to do this in one data step, where I can read from the data sets
and do the computation.

Could you help me with this? Thanks.



Regards, Max