Subscribe Bookmark RSS Feed

How do I create 3-way interactions in linear regression?

jswislar

Contributor

Joined:

Nov 17, 2016

I'm trying to create a model with a 3-way interaction term. I selected the variables and clicked on "cross" per JMP instructions for 2 or more variables. But JMP gives me the following warning:

 

"The model is missing an effect

X1*X2*X2*X1"

 

Which is especially odd since what I entered into the model, and what appears in the dialog box, is:

X1

X2

X3

X4

X5

X1*X2*X3

 

Any thoughts on how to resolve this?

Thanks

1 ACCEPTED SOLUTION

Accepted Solutions
markbailey

Staff

Joined:

Jun 23, 2011

Solution

If you want the term X1*X2*X3, then you must also include X1*X2, X1*X3, and X2*X3 to maintain the linear model hierarchy, even if these lower order terms are not significant.

Learn it once, use it forever!
2 REPLIES
markbailey

Staff

Joined:

Jun 23, 2011

Solution

If you want the term X1*X2*X3, then you must also include X1*X2, X1*X3, and X2*X3 to maintain the linear model hierarchy, even if these lower order terms are not significant.

Learn it once, use it forever!
jswislar

Contributor

Joined:

Nov 17, 2016

Ahhh, you're right! I was thinking that only the lower order terms needs to be included. Thanks!