Help with Decision Tree in SAS Enterprise Miner

[sollani]

Hi everyone,

I’m working on creating an interactive decision tree in SAS Enterprise Miner using a raw dataset (no train/validation split). However, I'm only seeing the top 4-5 variables for the first node split. Is there a way to increase this to consider at least 10 variables? Any suggestions would be greatly appreciated!

Thanks!

[ChrisNZ]

If my memory serves me right, you should look at the weight of evidence of the various variables before adding them.

There's no point having a more complex decision tree if it brings no benefit to the model.