I was trying to do something like this:
select nume_produs
from incasari
group by id
having count(nume_produs) = max(count(nume_produs));
but it doesn't work
I was trying to do something like this:
select nume_produs
from incasari
group by id
having count(nume_produs) = max(count(nume_produs));
but it doesn't work
Do a GROUP BY. Order by count descending. Fetch the first row (highest count) only.
select nume_produs, count(*) as cnt
from incasari
group by nume_produs
order by cnt desc
fetch first 1 row with ties
For the most common value in the column:
select num_produs
from (select nume_produs, count(*) as cnt,
row_number() over (order by count(*)) as seqnum
from incasari
group by nume_produs
) i
where seqnum = 1;
If you want multiple values in the event of duplicates, use rank()
instead of row_number()
.
If you want the most common value per id, then add partition by
:
select num_produs
from (select nume_produs, count(*) as cnt,
row_number() over (partition by id order by count(*)) as seqnum
from incasari
group by nume_produs
) i
where seqnum = 1;
SELECT `nume_produs`,
COUNT(`nume_produs`) AS `value_occurrence`
FROM `incasari`
GROUP BY `nume_produs`
ORDER BY `value_occurrence` DESC
LIMIT 1;
Increase 1 if you want to see the N most common values of the column.