cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Browse apps to extend the software in the new JMP Marketplace
Normalization GUI
hogi
Level XII

hogi_0-1694034101629.png

Purpose

The AddIn provides a GUI to automatically create normalized columns.

e.g.: weight normalized by weight of a typical 15year old boy/girl.

hogi_0-1694032907395.png

 

How to Use

Open the GUI via Cols/normalize , adjust all settings as you like, then click Apply.

 

GUI

hogi_0-1695233392815.png

 

 

Details

  • Click on Explain the GUI to get a tour of the available features.
  • Recall populates the GUI entries with the previous settings.

 

Columns can be used as:

  • columns to normalize: generate a new, normalized column for each column in the list
  • group by: different items are normalized separately, e.g. sex: different normalization values for male and female
  • subset selection: to pick a "reference subset", e.g. normalize by values of 15 year old students
    alternative applications: normalize by 0h data, by reference data, ...

 

Settings

  • normalize via divide or subtract
  • how to aggregate multiple values: median, mean, first value ...  new: N-1 = Diff (with grouping option)

 

The new column can be produced by (just) calculating the values or by generating a Formula Column.
Default setting: Formula which is calculated once and then deactivated.

 

System Configuration

Add-in developed and tested using the following system configuration:

 

Developed with JMP 17.1.

Tested with JMP 16-17.2.

Comments
hogi

This AddIn was triggered by a comment from @mia_stephens in a reply to a Wish List entry:

hogi_0-1694949217116.png

 

known restriction:
- subset selection doesn't work for columns with value labels (will be fixed with the next release)

 

to do: 
- add lag option

hogi
hogi
paulp

Hello, this GUI is useful, thank you.  Is it possible to add a robust scaler?  The most straightforward calculation is to subtract the median and divide by the interquartile range.  This should prove robust to outliers.

hogi

Great idea -
please follow this link and vote: Add normalization and robust statistical functions (and matrix functions) 
Then you get the Option directly via JMP.

 

For the GUI, I added an Option for IQR and robust normalization (-median & /IQR):

hogi_0-1733574246949.png

hogi

Unfortunately, 
it's not possible anymore to post updates for AddIns in the community:

hogi_0-1733564131896.png

If you are interested in updates, please send me a personal message.

I will also try to publish the latest version of the AddIn via the MarketPlace ...

Recommended Articles