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

least squares box-cox graphing improivement

JMP should support an improvement to the box-cox λ parameter when transforming the response in the least squares fit model, so that the reverse transformation is automatically applied to graphs. This would save messing around for hours withg the graph builder.

5 Comments
Status changed to: Acknowledged

@Grumpybaldprof - Thank you for your suggestion! We have captured your request and will take it under consideration.

SamGardner
Level VII

@Grumpybaldprof if you perform the fit on the Box Cox Transformed response (using the Refit with Transform or Replace with Transform) option under the Box-Cox transformation

SamGardner_0-1663943304990.png

and then you save the prediction formula for that fit (red triangle > Save Columns > Prediction Formula), that prediction formula has the appropriate inverse transform.  You can then use that column to create plots in Graph Builder. 

 

Does that meet your needs?  

 

SamGardner
Level VII
Status changed to: Needs Info
 
Grumpybaldprof
Level II

Thanks Sam,

Unfortunately no, this solutions does not help. The reason is simple, the transformation changes the axes values making them useless for biological interpretation. See screenshot below.

best wishes,
Paul
 

Grumpybaldprof_9-1678921110164.png

 

JMP® Pro 16.2.0

 

Grumpybaldprof
Level II

Hi Sam,

 

Ah now I see, but there seems to be no way in graph builder to have the raw points in the x axis, coded by colour for fetal sex and the 2 curves OVERLAID  from the predicted formula column coloured by sex. Please advise.

 

best wishes,

 

Paul

 

 

JMP® Pro 16.2.0