BookmarkSubscribeRSS Feed
kerenafK1
Calcite | Level 5

To whom may it concern,

 

I adjusted prior probabilities accordingly since my modeling data is non-event undersampled, using SAS Eminer 14.1

 

I noticed that ROC curve is flattened after that. I thought that AROC should not changed by over/undersampling, so I am very confused.

 

Can you kindly help me to understand?

 

Many thanks in advance,

Kere

2 REPLIES 2
WendyCzika
SAS Employee

Can you provide some more details about your flow, specifically the model nodes you are using?

kerenafK1
Calcite | Level 5

Hi!

 

Thank you for your kind reply. Please find the attachment containing the diagram overlook.


SAS Question.png

Catch up on SAS Innovate 2026

Nearly 200 sessions are now available on demand with the SAS Innovate Digital Pass.

Explore Now →
How to choose a machine learning algorithm

Use this tutorial as a handy guide to weigh the pros and cons of these commonly used machine learning algorithms.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 2 replies
  • 1723 views
  • 0 likes
  • 2 in conversation