Subscribe Bookmark RSS Feed

find possible correlation among variables


Community Member


Nov 14, 2017

Hello everyone,

I'm wondering if someone can help me with the analysis I have to do, since I'm new to this software.

In few words, I have a box let's say and the inputs to this box are A, B and C. The outputs are D, E, F, G, H. Each letter corresponds to a coloumn with many results. For example:

Inputs: A:100 200 300 400

            B: 123 456 789 123

Outputs: D:125 147 159 165

               E: 14 89 55 63

and so on.

I would like to find if there's a correlation among inputs and outputs and  which is the probability that a variable from the outputs comes from a variable of the inputs.


I would be really grateful if someone could answer.




Apr 26, 2012

A good place to start is with the Multivariate Platform (Analyze/Multivariate Methods/Multivariate ).  Just put all your variables into the Y role, and click OK.


One other thing that might be important for you.  I suppose you are starting out with data in some other table, not in JMP.  In the source program, select the range of data, with the column names on the first row, and copy this to the clipboard. Then in JMP do:  File/New/Data Table, Then without doing anything else, don't click on anything in the table, do this: Edit/Paste with Column names.   Now you should have a complete dataframe ready to roll with.