Assuming you are trying to remove all records from [state] which also
exist in [city]:
Proc SQL ;
create table only_A as
select * from state
where state.zip not in (select distinct zip from city)
order by zip ;
no great saving in code but you do not have to sort the two datasets as
you would before the merge
Dan wrote:
> How to use PROC SQL to replace following merge step:
>
> data only_A;
> merge state(in=a) city(in=b);
> by zip;
> if a and not b;
> run;
>
> Thanks.
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
