cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Check out the JMP® Marketplace featured Capability Explorer add-in
Choose Language Hide Translation Bar
VB
VB
Level I

Problem with the creation of a linear model

Aug 24, 2023 04:04 AM (419 views)

Hello,

I'm trying to create a linear model that predicts glucose per WD. Models are created for WDs ranging from 1 to 3, but no model is created for my WD=4. The curve for the prediction vs observed values  is not created.  The r-squared is 1, which is very strange. In the "effect tests" section, there is no Prob.> F. When I remove a quadratic term, I get a model. When I try to create a step-by-step model, I also have to remove a quadratic term to obtain a model. How can I solve the problem and have a model curve that appears directly?

 

VB_3-1692863789622.png

 

VB_2-1692863521014.png

 

VB_1-1692863500856.png

Thank you for your time

0 Kudos
1 REPLY 1
MRB3855
Super User MRB3855
Super User

Re: Problem with the creation of a linear model

Created: Aug 24, 2023 05:04 AM  |  Last Modified: Aug 24, 2023 2:10 AM (413 views)  |  Posted in reply to message from VB 08-24-2023

Hi @VB : It's hard to know what is going on without: seeing your data, knowing what your goals are, knowing how the data were acquired etc. But...R^2=1, no p-values etc. means you've fit the points exactly (connected the points exactly via your quadratic model).In that case, there is no error; so there can be no RMSE, p-values, leverage plots (observed vs predicted plot), etc.  Can you share more information? 

0 Kudos