1
$\begingroup$

From my knowledge about independent component analysis the number of channel or variables will not be affected by this algorithm and the number of independent components will be equal to number of variables. But when I tried to run FASTICA algorithm on EEG signal matrix which has 14 channels (or 14 variabls), the number of independent components was 9 not 14, which is confused me.

Does anyone have a clarification about this?

$\endgroup$
1
$\begingroup$

ICA can be seen as a dimension reduction technique similar to PCA. Although PCA consists of the eigenvectors of the covariance matrix i.e. second order statistics, ICA is not restricted and can contain higher order stat. The goal of ICA is to span the signal space in the most statistical independent vectors. So you might get a considerable dimension reduction.

$\endgroup$

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.