I am trying to evaluate the different ICA algorithms. To do that, one of the measure which I use, is to estimate the non-gaussianity using NegEntropy. I am trying to find a formula/function which can do this on time series EEG data. May I know how can I estimate it for a random variable 'x'.? And how it differs from the entropy of 'x'? Would appreciate any reference.

  • $\begingroup$ I think this question will be ruled off-topic because it is software specific. Nevertheless, are you just n need of kurtosis? See mathworks.com/matlabcentral/answers/… . $\endgroup$ – Mark L. Stone May 25 at 10:02
  • $\begingroup$ I have a matlab function to estimate Kurtosis using nl.mathworks.com/help/stats/kurtosis.html. However for negentropy it has not been easier to do. $\endgroup$ – Mari May 25 at 13:41
  • $\begingroup$ Sho9w us the formula you are trying to use? There is an approximation given at the link in your question. $\endgroup$ – Mark L. Stone May 25 at 14:58
  • $\begingroup$ Thanks. I have just added the reference page using which I estimate the Entropy. I am not sure whether i can use the entropy as an estimation for non-gaussianity instead of Negentropy. $\endgroup$ – Mari May 25 at 15:06
  • $\begingroup$ Negentropy is not entropy. I suggest you edit the question title and body to remove reference to MATLAB, and refocus the question on how to calculate Negentropy, and its use as a measure of non-gaussianity. $\endgroup$ – Mark L. Stone May 25 at 15:30

Your Answer

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

Browse other questions tagged or ask your own question.