Can I cross variables already added to a model (fit model)?

[#1]

I am hoping to use the macros (factorial to degree) cross on variables that have already been added to a model so that I don't have to find the columns from scratch (dealing with a very large data set) to be able to cross them. Any ideas? Also, I do not know much about programming.

[#2 · Accepted solution]

Hi,
This a functionality that most of us would love to have but it is not available, at least in JMP14.1.
One of the most efficient way to add interactions between variables in existing model is to:
1) run the model as-is
2) save the script: red top left triangle menu > Save script > To Script Window
3) In the Effects line of code, add the interactions you which to explore (make sure to use the exact syntax for variables as listed for the individual effects)
Example: :data1 * :data2, *:data1 * :data3....

[#3]

@Thierry_S

Alright, thank you! Hopefully, they add this feature soon!

[#4]

Perhaps this case would be easier if you  first isolate the data columns with the variables for the modeling using Tables > Subset.

[#5]

@Mark_Bailey 

Thank you for the idea!

[#6]

Maybe I'm not fully and correctly understanding your request, but couldn't you use the Edit function within the Effect Summary Fit Model report window? Selecting 'Edit' brings up a new Fit Model dialogue box and from there you can use the usual Fit Model model building macros, transforms, etc. So you can articulate your interaction effects and hit the 'Commit' button and your model report window will automatically update according to the model selections you've made.

Hope this helps?

[#7]

@P_Bartell 

Thank you for your reply. The trouble is that you have to re-select all the columns from the "select columns" box rather than being able to select them from the "effects" box once you are in edit mode if you would like to use Macros to cross them.