cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Choose Language Hide Translation Bar

Actual vs Predicted Residual Plot Apply to which types of fit personality

Hello!

 

I have been seeing some mentions of an actual Actual vs. predicted residuals plot to help with data analysis. however when analyzing my data though jmp, this plot does not come up for me nor is it an option. The jmp site mentions that this usually pops up by default. 

 

Could this plot be unavailable to me due to my use of the nominal logistic fit 

2 REPLIES 2
Thierry_S
Super User

Re: Actual vs Predicted Residual Plot Apply to which types of fit personality

Hi,

The Actual versus Predicted plot is only associated with models involving continuous Xs and Y. The confusion matrix in the ordinal logistic model delivers the same concept but for a categorical Y (i.e., Actual versus Predicted).

 

Best,

TS

Thierry R. Sornasse
Victor_G
Super User

Re: Actual vs Predicted Residual Plot Apply to which types of fit personality

Hi @MedianPuppy4982,

Depending on the personality of your linear model with continuous response(s), the "Actual Vs. Predicted" plot may not always appear.
You can make it appear by clicking on the red triangle, go to "Row Diagnostics" and click on "Actual Vs. Predicted plot" : https://www.jmp.com/support/help/en/17.2/#page/jmp/row-diagnostics.shtml#ww694286

For logistic model, as the response is not continuous but categorical, you won't have the same type of plot, but you can display the Confusion matrix showing the predicted classes Vs. Actual classes, available through the red triangle : https://www.jmp.com/support/help/en/17.2/#page/jmp/options-for-nominal-and-ordinal-fits.shtml#ww3151...

 

You can also make these graph appear by default by going into your JMP menu "Preferences", searching for these two graphs, and check them by default : https://www.jmp.com/support/help/en/17.2/#page/jmp/jmp-preferences.shtml


Hope this answer will help you,

Victor GUILLER
Scientific Expertise Engineer
L'Oréal - Data & Analytics