Quote:
Originally Posted by commodoreshrvk
Keep the data coming, I am putting it all into an Excel spreadsheet and will analyze it with a Discriminant Function Analysis (DFA) to look at where the group ?break? points are. Please only post in one thread. Right now I have 49 of your transfers in here and I can see the following pattern for the tiers:

I want to start by saying that I'm a mathematician, but statistics are definitely not my strong suit (I'm more of an algebraist). That said, I'm curious why you chose a DFA for your analysis. This problem doesn't seem to really satisfy the assumptions of a DFA model.
To be clear, I'd love to work with you to develop the most accurate model possible. Please don't interpret this as an attempt to discredit you, or to create a competing model.
My intuition as to how to solve this problem was as follows:
Assume that cryptic assigns a "value" to each commodity, and that if your "total value" is in a certain range, you will be assigned the crate corresponding to that category.
Set the value of an EDC to 1.
Make some assumptions, such as:
Value(EDC)<Value(Mk10Req)=Value(Rare Salvage) <=Value(Proto Salvage)=Value(Mk12Req)
Value(Mk11Req)=2*Value(Rare Salvage)
Value(EDC)<Value(Mk10tech)<=Value(Mk11Tech)<=Value (Mk11Tech)
Value(some level tech)=Value(Corresponding requisition)
Then treat the problem as a large system of linear equations. I'm not entirely sure how I'd deal with the fact that the "total value" is known only up to a range. I was planning on crossing that bridge when I came to it, but it's possible that it could prove insurmountable.
In any event, could you share your spreadsheet, so that I (and others) have access to all the same data as you do, in an easytoread format?