proc phreg with time varying covariates

  • Follow


Hello:

I am working on a problem using the Framingham data set. I am trying
to estimate the affect of diabetes on risk of death. I am using the
code below, but am getting the WARNING messages further below and no
output. Any insight into what is wrong?

Regards.

proc phreg data=temp;
  class sex/param=ref ref=first;
  model timedth*death(0) = x sex age1;
  array age(*) age1-age3;
  array diabetes(*) diabetes1-diabetes3;
  x = diabetes1;
  do i = 1 to 3;
     if age(i) ne "N" and timedth ge age(i) then x = diabetes(i);
  end;
run;

473  proc phreg data=temp;
474    class sex/param=ref ref=first;
475    model timedth*death(0) = x sex age1;
476    array age(*) age1-age3;
477    array diabetes(*) diabetes1-diabetes3;
478    x = diabetes1;
479    do i = 1 to 3;
480       if age(i) ne "N" and timedth ge age(i) then x = diabetes(i);
NOTE: Character value converted to numeric for argument 2 of '^='
operation
      at line 480 column 19.
481    end;
482  run;

WARNING: Cannot find a library containing subroutine DIABETES.
WARNING: Cannot find a library containing subroutine AGE.
NOTE: The SAS System stopped processing this step because of errors.
0
Reply njburn (1) 4/21/2010 11:29:31 PM


0 Replies
797 Views

(page loaded in 0.023 seconds)

Similiar Articles:













7/25/2012 8:51:50 AM


Reply: