Hi Q Experts,
I’m unsure of the best option in this instance so hopefully some of you can help.
Say I’ve a table with columns date, sym and size (small sample);
date sym size
---------------------
2018.02.01 `A 200
2018.02.01 `B 400
2018.02.02 `A 600
2018.02.02 `B 200
2018.02.05 `A 100
2018.02.05 `B 101
2018.02.06 `A 205
2018.02.06 `B 400
How would I reduce this table to only show me, say 5 records per date (in this case, the 5 highest sizes for each date, obviously I can sort those columns first if needs be).
The big question - would I be better using group or would fby be more efficient here and also _ why _ more importantly? - I was reading into fby on the following page (but was finding it confusing);
https://code.kx.com/wiki/Reference/fby
As always thanks!
Regards
Andie C