The first few answers are unidiomatic where I work. Analytical functions would be vastly preferable to self joins, especially in the case of the join with an in

The first few answers are unidiomatic where I work. Analytical functions would be vastly preferable to self joins, especially in the case of the join with an inequality that is proposed to compute running totals, which I assume would have horrible performance on large datasets