cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
  • New to JMP? Join us Sept. 23-24 for the Early User Edition of Discovery Summit, tailor-made for new users. Register now for free!
  • Use World Cup data to build models, explore spatial relationships, and create informative visualizations in JMP. Register. July 17, 2 pm US Eastern Time.
  • Your voice matters! Tell us how you prefer to receive JMP updates, so we can tailor our communication to your needs. Take short survey.

JMP Wish List

We want to hear your ideas for improving JMP. Share them here.
Choose Language Hide Translation Bar
0 Kudos

Custom inverse prediction for quadratic (and higher orders) models

In the fit curve platform the option to have a inverse prediction is not available for quadratic models. It would be nice to add it. Of course if possible to include cubic and others it would be top.

 

 

4 Comments
Status changed to: Acknowledged

Hi @friedrich, thank you for your suggestion! We have captured your request and will take it under consideration.

SamGardner
Level VII
Status changed to: Needs Info

@friedrich thank you for your suggestion.  I have a follow up question.  With a quadratic or higher order polynomial model, there is a possibility that that the fitted function will not be monotonic, and for a given Y value there could be two possible inverse solutions.  How would you propose that situation be handled?  

SamGardner_0-1669944697770.png

 

friedrich
Level II

@SamGardner , would it be possible to have both solutions in the output ? Or the user may choose before if he wants the result for the 1st or 2nd solution ?

QubitQueen100
Level I

I agree this would be helpful! I ran a standard curve with x as concentration and y as fluorescence. I now would like to predict concentration when given RFU. A 3rd-order polynomial fits my data well. However, I have to enter the fluorescence value one at a time using this workaround: https://community.jmp.com/t5/Discussions/inverse-prediction-using-quadratic-model/m-p/221830#M44262

 

I agree with friedrich ^ that all solutions could be output, or user could choose to display all real number solutions.