I have a table that has 3 columns: user_id, date, amount
. I need to find out on which date the amount reached 1 Million for the first time. The amount can go up or down on any given day.
I tried using partition by user_id order by date desc
but I can't figure out how to find the exact date on which it reached 1 Million for the first time. I am exploring lead, lag
functions. Any pointers would be appreciated.
CodePudding user response:
Try this:
select date,
sum(amount) as totalamount
from tablename
group by date
having totalamount>=1000000
order by date asc
limit 1
This would summarize the amount for each day and return 1 record where it reached 1M for the first time.
Sample result on SQL Fiddle.
And if you want it to be grouped for both date and user_id, add user_id in select and group by clauses.
select user_id, date,
sum(amount) as totalamount
from tablename
group by user_id,date
having totalamount>=1000000
order by date asc
limit 1
Example here.
CodePudding user response:
You may use conditional aggregation as the following:
select user_id,
min(case when amount >= 1000000 then date end) as expected_date
from table_name
group by user_id
And if you want to check where the amount reaches exactly 1M, use case when amount = 1000000 ...
If you meant that the amount is a cumulative amount over the increasing of date, then query will be:
select user_id,
min(case when cumulative_amount >= 1000000 then date end) as expected_date
from
(
select *,
sum(amount) over (partition by user_id order by date) cumulative_amount
from table_name
) T
group by user_id;