Hello,
I have a nominal response variable which can have two states 1 and 0 and 6 input variables (numeric, continuous). How can I find the best model to predict my response variable? This is probably a very broad question but I would like to understand the process of getting and building the best model for my response variable.
So some of the questions I have are the following:
1.) How do I know if I should use Partition or Fit Model?
2.) When I use Regression (Fit Model), how do I know which Macros to use? Full Factorial, Polynomial to Degree, etc.? To which degree? The higher the better?
3.) What about the personality? Should I use Nominal Logistic or Stepwise Regression?
As you can see I am a beginner in building predictive models and would like to understand the process or maybe there is guide that I can follow to build the best model for my data.
Thanks.
I suggest you start with some basics. These can be self guided or live webinars:
https://www.jmp.com/en_us/events/live-webinars/getting-started-with-jmp/getting-started-with-jmp-02apr2021.html
https://www.jmp.com/en_us/events/getting-started-with-jmp/doe-intro-kit.html
There are many ways to build models. Additive, subtractive. Regression from an existing data set or from a designed experiment. The models you are capable of estimating depend on how you got the data.
