In Incremental response model diagnostic chart or table, which one is predicting the variables to be used?
Train, Validate or Test? What would be the impact if the train is looking like a good model but Validate is not?
Thanks so much
Soma
Like other predictive modeling, the train and validation set are used for training in the IR node and the test set as hold-out data is only used for verifying the model accuracy at the end. If you see that the train is good but not validation, please try more options for training. For example, you can specify the Selection Method and set Validation Error or Cross Validation Error as the Selection Criterion; also the Two-Way interaction option can help capture nonlinear relationships if there is any in the data.
Thanks so much. These tweks helped my model getting desired results
Thanks
Hi!
I'm glad you found some useful info! If Ruiwen's reply was the exact solution to your problem, can you "Accept it as a solution"? Or if it was particularly helpful, feel free to "Like" it. This will help other community members who may run into the same issue know what worked.
Thanks!
Anna
Build your skills. Make connections. Enjoy creative freedom. Maybe change the world. Registration is now open through August 30th. Visit the SAS Hackathon homepage.
Register today!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.