Proc SQL

[KC_16]

Hi,

I have 2 datasets (work.one and work.two) and I need to append the data to one another.

However, if a record_id is in work.one and work.two I don't want it to merge, just those records_Id where there not in both tables so I have a unique list of accounts. 

How can this be done using PROC SQL please? I did think a UNION would work bit not 100%

[Satish_Parida · Accepted solution]

*selecting ONE-TWO UNION TWO-ONE;
proc sql;
create table result as
	select * from 
	(
		(select * from work.one where record_id not in (select distinct record_id from work.two)
		 union
		 select * from work.two where record_id not in (select distinct record_id from work.one)
		)
	)
;
quit;	

You had not made your self very clear, let us know if this helped.