Re: Identifying Important Variables in Your DataHello,
I found this video quite interesting and helpful! Thank you for that!
I would like to better understand the following: all the steps linked with interactions identification and assessment (with response screening and stepwise regression tools) are based on a linear assumption - is this correct? If yes, is there any suggestion to check for non-linear interactions?
Thank you in advance!
Helena
Re: Identifying Important Variables in Your DataNice data set and presentation. I will use it ! Thanks Nick
Re: Identifying Important Variables in Your Datagreat overview of rarely used but incredibly powerful platforms everyone should have in their tool case
Re: Identifying Important Variables in Your DataHi @HelenaG, I'm glad you enjoyed the video!
The identification of interactions is best done through domain expertise and a known scientific/physical understanding of the variables under analysis.
When interactions between variables are not well understood the "Response Screening" model personality in Fit Model works well at quantifying the strength between individual interaction terms and the response variable.
The "Stepwise Regression" model personality in Fit Model works well at identifying the best combination of terms (Ex: Main effects, interactions..) to use for a potential model.
The term "linear" model refers to how the formula of the model is structured but both the "Response Screening" & "Stepwise Regression" model personalities in Fit Model can also be used to explore non-linear relationships.
We can include non-linear effects and interactions into an analysis by adding polynomial terms.
In Fit Model you can explore any polynomial relationships (non-linear) by adding the corresponding term in the Model effects box (Ex: Quadratic Relationship (x1*x1), Cubic Relationship (x1*x1*x1), Quadratic Interaction (x1*x1*x2)...)
For more information on polynomials and how they can assist with understanding relationships among variables please watch the quick video below:
https://community.jmp.com/t5/Short-Videos/Fitting-Polynomial-Models/ta-p/271953
Re: Identifying Important Variables in Your DataHello again @nick_shelton,
Because of this tutorial, I am now using Screening Platform and I would like to ask you what are the settings of Bootstrap Forest behind the standard Screening Predictor Platform. In JMP Pro, there is a Bootstrap Forest Specification Window, and my question is what are the specifications for the bootstrap used in the screening platform (we can only define the number of trees). As a non-statistician, using Predictor Screening sounds less risky to me than using JMP Pro Bootstrap Platform; however, for scientific reporting, I would need to have further information on the bootstrap forest specifications behind this JMP platform.
Thank you very much in advance.
Helena G.
Re: Identifying Important Variables in Your DataHi @HelenaG
I'm happy to hear that you've been applying these concepts in your work 
The main difference between Predictor Screening (PS) and the Bootstrap Forest (BSF) in JMP Pro is that BSF provides options to control all of the hyperparameters that affect the results of the analysis.
As you've already noticed PS only allows for tuning of the "number of trees" parameter.
For more details related to the default hyperparameter settings used in PS, please forward your question to support@jmp.com which can provide some additional insights on how the method works.
Identifying Important Variables in Your Data
See how to:
- Find important factors in large data using Response Screening
- Examine FDR PValue Plot, FDR LogWorth by Effect Size, FDR LogWorth by RSquare
- Identify important variables using Predictor Screening, a decision-tree based method for screening many predictors for their ability to predict an outcome
- Identify unique important variables using variable clustering to group similar variables into representative groups
- Understand and examine how each cluster can be represented by a single component or variable to eliminate multicollinearity between variables
- Interpret Color Map on Correlations, Cluster summary, Cluster Members and Standard Components reports
- Identify interactions among unique and important factors using Fit Model - Response Screening
- Interpret FDR PValue Plot and FDR LogWorth by Rank Fraction reports using Graph Builder
- Identify the best subset of variables and interactions for building Regression Models
- Group similar variables into representative groups
- Interpret Stepwise Regression Control and Estimates reports
Response Screening to Help Identify Important Variables
Attached JMP Journal should work for all. Project file .zip file includes a JMP 15 Project Archive that requires unzipping to work with JMP 15 or later. See JMP archive documentation for help after you unzip the file.
Data Exploration and Visualization
