Can I use the best-subset selection procedure to select explanatory variables for nominal logistic regression analysis in JMP?

[#1]

Everyone
thank you always

Currently, I have to try all possible combinations (511 patterns) of 9 explanatory variables using the nominal logistic regression analysis method. It's difficult to do everything manually.
Is there a system in JMP that can select explanatory variables using a best-subset selection procedure like this?
It would be very helpful if you could let me know.

Sincerely.

[#2]

Hi @ErrorIguanodon1,

Welcome in the Community !

Maybe you could have a look at the Predictor Screening platform to identify which are the most important variables among the 9 explanatory variables ?

Also the Stepwise Regression Models available in the platform "Fit Model" may be helpful to explore the most influential variables on the response (a "data mining" approach to select the most influential terms in the model for observational data). 

Finally, some other models may be interesting to look at if you have JMP Pro to do some exploratory modeling, for example the Bootstrap Forest.

I hope this answer will help you,

[#3]

To answer your specific question, Select Analyze > Fit Model, then change the fitting Personality to Generalized Regression. It offers many methods for model selection, including the best subset.