Options

"multivariate correlation - groupwise correlation"

levietduclevietduc Member, University Professor Posts: 2 University Professor
edited May 2019 in Help

Hi everyone,

 

I want to compute the correlation coefficience between 2 or more groups of attributes. Mathematically, I have 2 groups: X and Y. X has 2 attributes (A, B), Y has 3 attributes (C, D, F). How can I compute R(X,Y)?

Tagged:

Answers

  • Options
    Telcontar120Telcontar120 Moderator, RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,635 Unicorn

    I do not believe there is a standard definition of correlation between asymmetrical groups of attributes, since there are several ways that it could be done based on the underlying correlations of the individual attributes.  Is there a reason you need to aggregate these at the group level rather than just looking at the correlation matrix for (A,B,C,D,F) which is easily produced in RapidMiner with the Correlation Matrix operator?

    Brian T.
    Lindon Ventures 
    Data Science Consulting from Certified RapidMiner Experts
Sign In or Register to comment.