Date: Wed, 20 Feb 2002 14:20:46 +0100
Reply-To: Dave Sorensen <Dave.Sorensen@JUR.KU.DK>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Dave Sorensen <Dave.Sorensen@JUR.KU.DK>
Subject: SV: coding a cumulative variable in univariate data
Content-Type: text/plain; charset="iso-8859-1"
Thanks for the suggestion. But the datastep's log file says: Variable
First.Subj is uninitialized. The result was that all subjects =1 on the new
variable (regardless of whether they had ever experienced the event).
Also, was I clear that the new variable should not =1 UNTIL the event has
occurred, but should then remain=1 thereafter.
Fra: Jim Groeneveld [mailto:J.Groeneveld@ITGroups.com]
Sendt: Wednesday, February 20, 2002 12:54 PM
Til: 'Dave Sorensen'; SAS-L@LISTSERV.UGA.EDU
Emne: RE: coding a cumulative variable in univariate data
Try the following program:
INPUT Subj Time X;
1 1 0
1 2 3
1 3 0
2 1 0
2 2 0
2 3 1
PROC SORT DATA=basic; BY Subj Time; RUN;
SET basic; BY Subj Time;
IF (First.Subj) THEN EverX = 0;
IF (X GT 0) THEN EverX = 1;
PROC PRINT DATA=cumul; RUN;
Regards - Jim.
Y. (Jim) Groeneveld, MSc IMRO TRAMARKO tel. +31 412 407 070
senior statistician, P.O. Box 1 fax. +31 412 407 080
senior data manager 5350 AA BERGHEM IMRO TRAMARKO: a CRO
J.Groeneveld@ITGroups.com the Netherlands in clinical research
As time elapses it is getting later.
Notice of confidentiality: this e-mail may contain confidential information
intended for the addressed recipient only.
If you have received this e-mail in error please delete this e-mail and
please notify the sender so that proper delivery
can be arranged.
> -----Original Message-----
> From: Dave Sorensen [SMTP:Dave.Sorensen@JUR.KU.DK]
> Sent: Wednesday, February 20, 2002 12:27 PM
> To: SAS-L@LISTSERV.UGA.EDU
> Subject: coding a cumulative variable in univariate data
> Dear SAS-L,
> I have a univariate data set where each subject-wave is a separate
> I have a variable called X. I would like to create a dummy variable
> EVERx which=1 if X is, or has ever been greater than 0. The result should
> look like this.
> SUBJ TIME X EVERx
> 1 1 0 0
> 1 2 3 1
> 1 3 0 1
> 2 1 0 0
> 2 2 0 0
> 2 3 1 1
> Is there a simple way to code this with the data in univariate format, or
> I need to transform the data to multivariate format, code EVERx wave to
> wave, and then transform it back to univariate format?
> Thanks in advance!
> Dave Sorensen