I'd like to create a data set containing only observations that have a record where FIRST=1 within each ID group. In the example below, I'd like to keep all observations except those where ID=60, since this ID group does not have any records where FIRST=1.
HAVE:
ID FIRST
7 0
7 1
18 0
18 1
56 0
56 0
56 0
56 1
60 0
60 0
76 0
76 0
76 1
WANT:
ID FIRST
7 0
7 1
18 0
18 1
56 0
56 0
56 0
56 1
76 0
76 0
76 1
This seems like it should be very simple coding, but I'm spinning my wheels here!
Thank you,
Stephanie
Alternatively via data step
proc sort data=have;
by id descending first;
run;
data want;
set have;
by id descending first;
retain test;
if first.id then test=first;
if test>0;
drop test;
run;
Thanks,
Jag
Are you looking for a data step or SQL solution? What hasn't worked for you?
I am much more familiar/comfortable with coding using data step.
The response by Jag below worked for me.
proc sql;
create table want as
select * from have
group by id
having sum(first)>0
order by id, first;
quit;
Alternatively via data step
proc sort data=have;
by id descending first;
run;
data want;
set have;
by id descending first;
retain test;
if first.id then test=first;
if test>0;
drop test;
run;
Thanks,
Jag
Thank you!
It's finally time to hack! Remember to visit the SAS Hacker's Hub regularly for news and updates.
Learn how use the CAT functions in SAS to join values from multiple variables into a single value.
Find more tutorials on the SAS Users YouTube channel.
Ready to level-up your skills? Choose your own adventure.