topic Re: base sas in SAS Programming

base sas

knveraraju91 — Thu, 19 May 2016 21:18:46 GMT

Dear ,

Ihave two large SDTM data sets. I need to find the number of subjid which are present in one datset and not in other dataset and viceversa. Is there any code that I can use.Thanks.

Thanks

Re: base sas

ChrisNZ — Thu, 19 May 2016 22:30:10 GMT

If you can index your tables by subjid, you can then do this:



data WANT;
  merge TAB1(keep=SUBJID in=A) 
             TAB2(keep=SUBJID in=B);
  by SUBJID;
  if first.SUBJID and not(A and B);
  if A then SOURCE='TAB1'; else SOURCE='TAB2';
run;

Re: base sas

PGStats — Fri, 20 May 2016 02:42:13 GMT

Or use SQL

proc sql;
create table want as
select "TAB1" as source, subjid from TAB1 where subjid not in (select subjid from TAB2)
union all
select "TAB2" as source, subjid from TAB2 where subjid not in (select subjid from TAB1);
quit;

Re: base sas

FreelanceReinh — Fri, 20 May 2016 08:28:25 GMT

If you're only interested in how many distinct, non-missing SUBJIDs there are in the first dataset and not in the second and vice versa:

proc sql;
select count(subjid) as only_in_tab1
from ((select subjid from tab1)
      except
      (select subjid from tab2));

select count(subjid) as only_in_tab2
from ((select subjid from tab2)
      except
      (select subjid from tab1));
quit;

Re: base sas

knveraraju91 — Wed, 25 May 2016 16:26:13 GMT

Thank you all