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; } } }

Can settings be maintained in JMP?

winfriedkoch0

The prediction profiler in JMP can be effectively used to visualize the dependence of the target variable from independent variables in the model. When I do this, I usually spend effort to improve the different compartments of the prediction profiles e.g. by optimizing scales and axis settings to make the prediction profiles as informative as possible.

Unfortunately all these settings are immediately lost e.g. when I remove one variable from the model. Of course, I can copy axis settings from the previous plot for one axis after the other but this is still time consuming for more extensive models.

Is there a possibility to maintain settings in the above and similar situations?

Accepted answers

sseligman

Hi Winfried,

I see the behavior you described where your customized axis label orientation in the Profiler reverts back to the default when removing an effect from the model. I will communicate this to our development team to see if we can address it in a future version.

Sarah

sseligman

Hi Winfried,

As Jeff mentioned above, if you change any axis settings, you can right click on the axis and select "Save to Column Property." That will save those axis settings to the column in the data table and thus they will be maintained in the Profiler when you remove an effect from the model. If you have other customizations that do not persist, please share them with us and we can try to address those as well.

Thanks,

Sarah

All comments

txnelson

There does not seem to be many options within the Preferences for the Profilers, that would allow for setting display options. However, since the profilers are made up with standard Objects, such as FrameBox()s and AxisBox()s, then it should be possible to reset the output through a script.

Other individuals on the Community Discussion Forum may have better input on this.

winfriedkoch0

Hi Jim,

Thanks for your feedback. I conclude from your answer that there is no easy solution for the issue indicated by me. As a consequence I intend including a corresponding request in the JMP wish list, because I think that this would improve the work flow in JMP.

Jeff_Perkinson

Can you describe more about how you're adding or removing a variable from the model? When I use the Effects Summary section at the top of the report the Profiler axis settings are retained.

Additionally, after making your adjustments to axes, you can right click on the axis and choose "Save to Column Property". This will save your axis settings to a property in that column. Those settings will be used for every axis with that column.

winfriedkoch0

Hi Jeff,

Thanks for your response.

Yes, I use the Effect Summary section at the top of the report to remove or add variables. To understand my issue, please look at the x-axis of your example: the label orientation at the x-axis of all your variables is initially vertical. Please change the label orientation of all variables to horizontal (=the intended orientation), then remove one variable. In my environment the label orientation of the remaining 3 variables then has switched back to vertical again; this is also the case if I save the horizontal setting to column property.

Best regards

Winfried

sseligman

Hi Winfried,

Sarah

winfriedkoch0

Hi Sarah,

Thanks for your understanding of the issue and your communication to the Development Team. Please note that I used "label orientation" and "removing an effect from the model" just as examples. Currently any customized axis settings in the Prediction Profiler are reverted back to the default by most changes of the model e.g. also by column switching.

Winfried