data step query

[rahulks]

If there is a condition,in which we have to take only those observations which are present in the same variable of different dataset, how we will write the query?

[PGStats]

Something like this :

proc sql;

create table A_and_B as

select * from A

union all corresponding

select * from B;

quit;

selects only variables in table A and B which have the same name and type. Other variables not in common are dropped.

PG

[rahulks]

thnx for the reply, but this query will not help me...

actually I works for the bank, and I have to put a query in a dataset which implies that, if account number is in accounts then table, then only account number will be added to the table on which I am applying my query.

[PGStats]

A small example (input and output data) would help us to understand what you want.