cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
  • DownloadSemiconductor Toolkit: Tools to create wafer maps, add wafer geometry to graphics, explore die defects & compare wafers.
  • Discovery Summit 2026: Early User Edition - September 23-24.Register. It's free.
  • Use JMP Clinical with CDISC-compliant data. Review studies, ID safety and efficacy signals & communicate findings with interactive graphics. Register to see how. Aug. 27, 2 pm US ET.

Discussions

Solve problems, and share tips and tricks with other JMP users.
Choose Language Hide Translation Bar
abmayfield
Level VI

Order of predictors under profiler

Hello all, 

    I have been gravitating more and more to the profiler and was wondering one thing: when you do a desirability analysis, the order of the predictors changes. It doesn't go to alphabetical order, nor does it follow the order in which the predictors are listed in the JMP data table. Instead, are they getting ordered with respect to something else, such as their impact on the model or their desirabilities? I did see that I can change the order under "Appearance" to basically override the order created by JMP, but I suspect the X's are ordered in a logical way! Does anyone know?

Anderson B. Mayfield
1 ACCEPTED SOLUTION

Accepted Solutions
abmayfield
Level VI

Re: Order of predictors under profiler

Never mind, I figured it out! But if anyone out there is reading this and had the same question, the answer is: I had done an "assess variable importance" analysis, and the predictors were automatically re-ordered to match their relative influence on the model, i.e., from most to least important.

Anderson B. Mayfield

View solution in original post

1 REPLY 1
abmayfield
Level VI

Re: Order of predictors under profiler

Never mind, I figured it out! But if anyone out there is reading this and had the same question, the answer is: I had done an "assess variable importance" analysis, and the predictors were automatically re-ordered to match their relative influence on the model, i.e., from most to least important.

Anderson B. Mayfield

Recommended Articles