cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Try the Materials Informatics Toolkit, which is designed to easily handle SMILES data. This and other helpful add-ins are available in the JMP® Marketplace
Choose Language Hide Translation Bar
Alittle
Alittle
Level I

Creating repeatable views for DoE Analysis

Created: Mar 16, 2022 12:10 PM  |  Last Modified: Jun 10, 2023 4:45 PM (532 views)

I'm noticing a few things in JMP14 that are different from my prior experience. Please advise if there is a way to create settings that I can use across my analyses.

1. The prediction profiler is not reflecting the values in the columns. I have to manually change the range displayed. I have tried changing and saving to columns but this doesn't seem to make anything change as when I run another analysis the column information has reverted to -1 to 1.

2. The default features displayed for the analysis are not those that I would like to see. I would like to hide the student's residual and display the summary of fit. For each analysis that I perform, I have to change what is displayed. 

Making these changes is annoying but doable when it is a small analysis but when it is more complicated (i.e. many outputs), this is time consuming. Also, if I re-run the analysis for any reason I have to repeat this effort. Note: When I save the script, this does save but the saved script may not cover the analysis I want to perform.

 

Thank you for your help.

0 Kudos
1 REPLY 1
statman
Super User statman
Super User

Re: Creating repeatable views for DoE Analysis

Mar 16, 2022 02:05 PM (520 views)  |  Posted in reply to message from Alittle 03-16-2022

Welcome to the community.  I have the following thoughts and advise:

1. The analysis of experiments is greatly facilitated by coding of the variable levels (this essentially normalizes the coefficients which is what we are analyzing).  You can use actual values of the factors and the output model should contain the coefficients based on actual values (even thought behind the scenes the levels get coded).  What are you doing when you run fit model?  If you are analyzing the coded levels, the prediction profile will show those values.  If you are analyzing the columns with actual values, the prediction profiler should provide the results in actual values.

2.yes, you can set preferences in JMP to accomplish what you want.  You didn't tell us what OS you are running and there may be slight variations across those, but what you want is to select File→Preferences.  Then navigate to Platforms>Fit Least Squares.  There are many options to select or de-select.

"All models are wrong, some are useful" G.E.P. Box