How do I apply a Bootstrap Forest model for future predictions on new data?

[#1]

I am new to JMP and after completing the online Statistical Thinking for Industrial Problem Solving course, I have a few questions. How do I save a particular optimized model of say a Bootstrap Forest (after determining it has a desirable R-squared and MSE) and then apply it to predict results on a new test data? How do I see the newly predicted results? This is the whole essence of predictive modelling. Is it possible to see the residual plot of my selected Bootstrap Forest model? The online course only presented how to evaluate a model's performance, without showing how to apply the models for future predictions.

Thank you.

Highest regards,

Paul

[#2 · Accepted solution]

Hi Paul,

You can save the prediction formula for the selected model to the data table. This formula can be used to predict the outcome variable for new observations. There is brief discussion of how to save prediction formulas scattered throughout the predictive modeling module. 

An alternative, in JMP Pro, is to publish your formula to the formula depot. The Formula Depot has additional functionality for deploying your model. 

You can search for Formula Depot in the JMP help at jmp.com/help to see how to use the Formula Depot. There are also some nice live and recorded webinars that you might want to take a look at.

Data Modeling & Predictive Modeling

Live webinar: https://www.jmp.com/en_us/events/live-webcasts/mastering-jmp/data-mining-predictive-modeling-16aug20...

https://www.jmp.com/en_us/events/ondemand/mastering-jmp/data-mining-predictive-analytics2.html

Building Better Models – Understanding Principles and Implementing Models 

https://www.jmp.com/en_us/events/ondemand/mastering-jmp/building-better-models-part1-understanding-p...

https://www.jmp.com/en_us/events/ondemand/mastering-jmp/building-better-models-part2-implementing-mo...

I hope this helps.

Mia