nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

526326_en-US

liliconcom

Re: least squares box-cox graphing improivement - Status changed to: Acknowledged

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

Re: least squares box-cox graphing improivement - Status changed to: Needs Info

Re: least squares box-cox graphing improivement

@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

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?

Re: least squares box-cox graphing improivement

Hello Sam,

Thank you for the quick response.

I am afraid this does not help, if you look at the screenshot below you will see why, original data on left, replace with transform on right - look at the units - it has made the fetal weights way off what they actually are - the graph is therefore uninterpretable.

Please advise.Thanks, Paul

[_SID_]
Version=16.0
Revision=16.0
Platform_long=Mac OS X 64-bit
Platform_short=m64
Order=9CSD6J
License_prod=JMP
Setnumid=70275702
SID_schema=2
SID_header=JMP 16.0
[_Info_]
[_FileData_]
$_Filename=JMP.per
$_Path=.\
$_Date=
$_Time=
$_Stream=-------------------- BEGIN --------------------
/* Do NOT modify the contents of this file. Version: 16.0 */
 Platform=Macintosh
 Product=JMPPRO      
 Release=16.0.x
 LType=SiteLicense
 EMode=Full
 SiteID=70275702
 MaxNUsers=1
 Starts=01JAN2001
 Expires=29SEP2023
 Administrator= 
 Organization=UNIVERSITY OF ABERDEEN
 Password1=dcfa5ea056d5bc3267034d8da5605a07
 Department=
$_Stream=--------------------- END ---------------------
$_Filename=tletter.txt
$_Path=.\
$_Date=
$_Time=
$_Stream=-------------------- BEGIN --------------------
Mac OS X 64-bit
JMP Pro 29SEP2023
$_Stream=--------------------- END ---------------------
[_EOF_]

Re: least squares box-cox graphing improivement

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

JMP® Pro 16.2.0

Re: least squares box-cox graphing improivement

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

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.