1
$\begingroup$

Are there mutual-information based correlation measures that are robust to compositional data?

It is my belief that many of these methods (e.g., distance correlation, transfer entropy, MIC, ...) may assume euclidean probability space.

$\endgroup$
1
$\begingroup$

Columbia University's Statistics Dept made the academic year 2013-2014 a year of focus on measures of dependence. In April-May 2014, a workshop was held that brought together the top academics doing work in this field including the Reshef Brothers (MIC), Gabor Szekely (distance correlations), Subhadeep Mukhopadhay to name a few. The modal space analyzed was the RKHS or Reproducing Kernel Hilbert Space, not euclidean.

Here's a link to the program that includes many pdfs from the presentations.

http://dependence2013.wikischolars.columbia.edu/Nonparametric+measures+of+dependence+workshop

$\endgroup$
1
  • $\begingroup$ @justin Did you lose interest in your question? $\endgroup$ Oct 28 '15 at 10:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.