cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Browse apps to extend the software in the new JMP Marketplace
Choose Language Hide Translation Bar
Mathej01
Level III

How to add more response variables?

I made a custom design with a set of response variables. Then I ran the experiments. But now i want to add some extra responses which are relevant for the DOE. Is it possible? if yes how to do it?

1 ACCEPTED SOLUTION

Accepted Solutions

Re: How to add more response variables?

I will try to help you with your request.

 

Your design is independent from the amount of responses you measure. 

Just add new columns to your design by double clicking into a new column field and track your results. 

 

If you need to add a goal for every new response check out the "response limit" column property for you new columns and add a goal like "maximize" there. 

 

When you run your model script in the upper left hand corner add your new response columns into the Y field.

 

Hope that helps.  

View solution in original post

3 REPLIES 3

Re: How to add more response variables?

I will try to help you with your request.

 

Your design is independent from the amount of responses you measure. 

Just add new columns to your design by double clicking into a new column field and track your results. 

 

If you need to add a goal for every new response check out the "response limit" column property for you new columns and add a goal like "maximize" there. 

 

When you run your model script in the upper left hand corner add your new response columns into the Y field.

 

Hope that helps.  

Mathej01
Level III

Re: How to add more response variables?

Thank you Jonas.

Note: The reason why i posted this question was to also  make sure if the number and charachter responses have any impact in design creation. And thank you sop much for making that clear as well. 

 

Re: How to add more response variables?

Happy to help.

 

Just be aware that binary or categorical responses will contain way less information than a continuous numerical value. So there is a high chance that your models will not fit as well as they do with continuous response values. 

You will also need other regression models like nominal logistic regression for these responses. The fit model platform will automatically highlight the regression models you need.