Proc SQL: Count unique using case

ywon111 — Sun, 10 May 2020 01:16:09 GMT

Trying to create the table below, i.e. adding another condition in the codes where if there is same Event_date then only Count_done as 1.

PROC SQL;
Create Table want1 as select
want.*, source.result_date,
case
when standard_days ge result_date and Event_date le result_date then '1'
else '0'

end
as Count_done
from want left join source on want.ID = source.ID;
Quit;

Current code produces this:

ID	Event_date	Count_done
123456	17/07/2018	1
123456	17/07/2018	1
234567	12/07/2018	1

This is what I want:

ID	Event_date	Count_done
123456	17/07/2018	1
123456	17/07/2018	0
234567	12/07/2018	1

Thanks in advance

Re: Proc SQL: Count unique using case

Tom — Sun, 10 May 2020 02:19:16 GMT

Why would you want to do that in SQL?

data want;
  set have;
  by id event_date;
  count_done = first.event_date;
run;

Re: Proc SQL: Count unique using case

ChrisNZ — Sun, 10 May 2020 05:31:59 GMT

You are not showing all the fields you process.

topic Re: Proc SQL: Count unique using case in SAS Programming

Proc SQL: Count unique using case

Re: Proc SQL: Count unique using case

Re: Proc SQL: Count unique using case