Date:         Tue, 10 Apr 2001 10:00:41 +0200
Reply-To:     Gijs Dekkers <gd@PLAN.BE>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Gijs Dekkers <gd@PLAN.BE>
Organization: planbureau / bureau de plan
Subject:      2nd question on PROC CORR
Content-Type: text/plain; charset=us-ascii

Dear fellow SAS-users,

Another question on PROC CORR, I'm afraid.

I have two variables SCV1 and SCV2, and I use proc corr to derive the
correlation coefficient. But I want to destillate this one figure from
the outputfile generated by this procedure.

I have used the following datasteps, but this is not good, because it
does not always work. Sometimes, the variables COL1 (see below) is
missing.

So, the question is: how can I destillate just the one correlation
coefficient from the output-file of the PROC CORR-statement?

Any help would be greatly appreciated.

Gijs Dekkers

===
Proc Corr data=alles out=f3 ;
 var SCV1 SCV2 ;
run ;

data f3 ;
  set f3 ;
  drop SCV2 ;
  if (_NAME_='SCV2') then output ;        /* _name_ <> varname.
Otherwise corr=1 */

PROC TRANSPOSE OUT=TEMP;
   BY _NAME_ ;

PROC SORT OUT=TEMP;
   BY _NAME_;

PROC TRANSPOSE DATA=TEMP OUT=FINAL;
   VAR COL1;            /* this one is not always there */
   BY _NAME_;
RUN;

data corrl ;
  set final ;
  CRR=COL1 ;
  drop COL1 ;


--
dr. Gijs Dekkers
Federaal Planbureau
Algemene Directie
Kunstlaan 47-49
B 1000 Brussel
++32/(0)2/5077413
fax  7373