Join and select column with max value - sql

Question

I am working on two tables in Oracle, TABLE_A containing unique IDs and Balance (number). TABLE_B shows, for the same IDs, the specific transactions and contains the following fields: IDs (not unique), BAL, Sequence_number.
I want to check that TABLE_A.Balance is always equal to TABLE_B.Balance having Max(Sequence_number).
So I expect to have just one row for each ID.
I've tried the following, yet it does not return a unique row for each ID but multiples. Why is that?
Select a.ID, a.Balance,b.Balance, b.sequence_number From TABLE_A a Inner join (select ID, Balance, max(sequence_number) as sequence_number from TABLE_B group by ID, Balance) b On a.ID = B.ID Group by a.ID, a.Balance, b.Balance, b.sequence_number
TABLE_A
ID_______Balance
1_______10
2_______15
3_______50
TABLE_B
ID____Balance____Sequence_number
1_______19_______1
1_______75_______2
1_______10_______3
2_______39_______1
2_______15_______2
3_______120_______1
3_______89_______2
3_______57_______3
3_______50_______4

Answer 1

You can use window functions:
select a.*, b.balance
from a left join
(select b.*,
row_number() over (partition by id order by sequence_number desc) as seqnum
from b
) b
on b.id = a.id and b.seqnum = 1;
I'm not quite sure what you want to compare, but this returns every row in a with the row in b that has the highest sequence number.

Answer 2

You can use row_number()over() window function to get the row having ID wise highest sequence_number.
Schema:
create table TABLE_A(ID int, Balance int);
insert into TABLE_A values(1,10);
insert into TABLE_A values(2,15);
insert into TABLE_A values(3,50);
create table TABLE_B (ID int , Balance int,Sequence_number int);
insert into TABLE_B values(1,19,1);
insert into TABLE_B values(1,75,2);
insert into TABLE_B values(1,10,3);
insert into TABLE_B values(2,39,1);
insert into TABLE_B values(2,15,2);
insert into TABLE_B values(3,120,1);
insert into TABLE_B values(3,89,2);
insert into TABLE_B values(3,57,3);
insert into TABLE_B values(3,50,4);
GO
Query:
with cte as
(
select id, balance, sequence_number, row_number()over (partition by id order by sequence_number desc)rnk from table_b
)
Select a.ID, a.Balance,b.Balance, b.sequence_number From TABLE_A a
Inner join cte b on a.id=b.id and rnk=1;
GO
Output:
ID
Balance
Balance
sequence_number
1
10
10
3
2
15
15
2
3
50
50
4
db<>fiddle here