Hi All,
I have a long data set of historical bond trades with columns representing characteristics for each (User, type, the bond id (cusip), company name, ticker, maturity, rating etc..)
I am trying to think of the best way to be able to enter a certain user or cusip, and having it return with who would most likely purchase that cusip; Or entering the user, and and have it return cusips they would most likely buy.
I have thought about using an evaluation/decision making matrix to give bonds scores or ratings based off of their values given on their row. However wouldn't this be tough due to the change in the cusip or user entry?
My goal is to have a function that will basically return all values in a ranked order of likeliness, "if this person buys X, then he also might like Y"
Any help or recommendations would be greatly appreciated
Bookmarks