[Eeglablist] Using PCA before ICA decomposition? (Chang Gu) (Chang Gu)

[stauros dimitriadis]

Dear Chang 

I suggest to re-reference your data to the averaged-reference for the reasons you describe. Especially, you recorded brain activity with a high number of electrodes and the averaged-reference is  more recommended
compared to a lower number of sensors. I suggest to adopt zscore to your analysis.

If you want to find extreme channels that differ compared to the rest regarding their amplitudes (mVolt), you can use the last output of
the princomp function,t2, is Hotelling'sT2, a statistical
measure of the multivariate distance of each observation from the center of the
data set. This is an analytical way to find the most extreme points in the
data.
[st2, index] =
sort(t2,'descend'); % Sort in descending order.
extreme = index(1)




Best regards
Stavros Dimitriadis
http://users.auth.gr/~stdimitr/index.html 

1)Electronics Laboratory, Department of Physics, University of Patras
2)Artificial Intelligence Information Analysis lab Department of Informatics
Aristotle University of Thessaloniki
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20120122/051285be/attachment.html>