cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Choose Language Hide Translation Bar

Adding the bootstrap forest feature to my JMP

I have a question about the features in the student version of JMP. I wanted to run a bootstrap forest analysis however, I am not seeing that feature available in my version. Would someone be able to assist me with obtaining that analysis feature?

2 REPLIES 2

Re: Adding the bootstrap forest feature to my JMP

Bootstrap Forest is a platform available in JMP Pro only. Is your student version JMP or JMP Pro (Help...About JMP)? The Predictor Screening platform in standard JMP is a subset of the Bootstrap Forest, specifically the ranking of column contributions portion of the platform.

Victor_G
Super User

Re: Adding the bootstrap forest feature to my JMP

Hi @PivotalRaccoon6,

 

Welcome in the Community !

 

Just to add some infos on the excellent response from @Jed_Campbell.

I'm also not familiar with the Student licensing, so depending if the licence is based on JMP or JMP pro, you might want to use Bootstrap Forest (available in JMP Pro only) or Predictor Screening (available in JMP and JMP Pro).

  • Bootstrap Forest can be found in the menu "Analyze", "Predictive Modeling", "Bootstrap Forest". The JMP Help pages dedicated to this Bootstrap Forest (jmp.com) model might be helpful to better understand how this type of model works and how to  Launch the Bootstrap Forest Platform (jmp.com).
  • Predictor Screening (JMP) can be found in the menu "Analyze", "Screening", "Predictor Screening". The JMP Help pages dedicated to this platform are quite helpful to understand : Predictor Screening (jmp.com) Basically, Predictor Screening platform launches a bootstrap forest (with 100 decision trees) to identify potential predictors of your response. A ranking of the column contributions (features importance) is then displayed as a result, but you can't use this platform to create predictive model/formula. You can see an example here : Example of Predictor Screening (jmp.com).

 

I hope these additional comments will help you find the platform and/or have other alternatives to use in your topic. If you need more guidance/help on the modeling with tree-based models (or others), feel free to open a new topic, and JMP users will help you.

Victor GUILLER
Scientific Expertise Engineer
L'Oréal - Data & Analytics