BookmarkSubscribeRSS Feed
harsh0404
Fluorite | Level 6

I am trying to figure out if i should do any oversampling or under-sampling in data-set where my event of non interest if very small. So i am trying to predict churn of customers. People who churned constitute 80%. the other 20% did not churn. My goal is to score people with churn probabilty (would use that in calculation of lifetime value). 

What should I do? ( I know how to do things in sas code, and I have eg and em as well, so how to do part in software i can do)

1 REPLY 1
Ksharp
Super User

20% bad event rate is good enough, no need to oversample .

hackathon24-white-horiz.png

2025 SAS Hackathon: There is still time!

Good news: We've extended SAS Hackathon registration until Sept. 12, so you still have time to be part of our biggest event yet – our five-year anniversary!

Register 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
  • 1 reply
  • 923 views
  • 1 like
  • 2 in conversation