Let's say I have 10.000 booking customer profiles. These profiles have the following variables:
- Duration (days to go on holiday)
- Destination (maybe Brazil)
- People_amount (how many go with)
- Takeoff (date when they want to leave)
I want to pass 1.000 (10%) of them through a booking engine for pricing, but to develop an independent insights analysis I have to (as much as I can) evenly distribute the characteristics of the profiles. E.g. if all the the profiles have 3 varieties of People_amount (1, 2 and 3), ultimately I would like to have a selection within the 10% of 33,33% with People_amount = 1, 33,33% with People_amount = 2 and 33,33% with People_amount = 3.
But...
Because the profile set isn't evenly distributed (for example 70% of all profiles consist of People_amount = 1) I can't figure out how to maybe find/create a sort of loop (or something else) that fills up the SELECT by all varieties within that characteristic untill 1 is exhausted and goes further with the rest.
Maybe an example on how I would like to fill up the 10% sample of my 10k profiles:
Profile_id People_amount Profile_id People_amount 1 1 1 1 2 1 5 2 3 1 8 3 4 1 --> Filling the sample by even distribution 2 1 5 2 of available profile characteristics 6 2 6 2 9 3 7 2 3 1 8 3 7 2 9 3 4 1
Hope you can help!