1

I have a set of data (7 variables) that are used as input to ultimately making a decision to approve or not approve something. I've generated two discriminant functions that I'm happy with - one for Approval and one for Not Approval, but now realize I'm missing a good understanding of the last step.

Taking a new record, I obviously solve the discriminant function for Approval and get a value, and the same thing for the Not Approve function. But how do I actually decide whether that record should be approved or not? I feel like this is an obvious question, but the textbook I'm referencing and the articles I've read don't directly address this.

Andy
  • 18,070
  • 20
  • 77
  • 100
Tee
  • 11
  • 1
  • If you are speaking about how observations are being classified by the extracted discriminant functions - please search this site `discriminant analysis classification`. There have been a number of answers by different participants (one my is this: http://stats.stackexchange.com/a/31384/3277). – ttnphns Jun 11 '15 at 19:25

0 Answers0