BookmarkSubscribeRSS Feed
jaycette
Calcite | Level 5
Hi,

I'm trying to build a decision tree in Enterprise Miner, but the splits aren't what I was expecting. I believe this is because my target variable is a ratio that ideally I want to be close to 1, but the tree is trying to find how to maximize the target. How can I change that to help me find more appropriate splits?

The ratio is simply variableA/variable B and was calculated in excel before inputting the data into sas, but I was also wondering if there was a way to have sas create the ratio and then use that as a target.

Thanks!
1 REPLY 1
WendyCzika
SAS Employee

I'm not sure I understand your statement: "I believe this is because my target variable is a ratio that ideally I want to be close to 1, but the tree is trying to find how to maximize the target."   The Decision Tree is trying to come up with rules based on your inputs for predicting your target, that is, the splitting rules that create the most significant difference in the target values of the resulting branches.

SAS Innovate 2025: Call for Content

Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 16. Read more here about why you should contribute and what is in it for you!

Submit your idea!

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
  • 659 views
  • 0 likes
  • 2 in conversation