[Eeglablist] EEG time series

[ali zahedi]

Dear list,

I have 10 EEG and 10 fMRI datasets recorded from 10 subjects.  I want to
apply a statistic method, Canonical Correlation Analysis (CCA) on my EEG
and fMRI data.
To do so, I used the sliding window of 1 to extract some features from all
EEG channels for each dataset. The total number of windows in each channel
is 80.
For CCA, the rows of the matrix must be independent, so here subjects are
independent observation and are arranged in a matrix as 10 rows and the
columns are features from each window of each channel.

My question is that are the windows from which I extarcted features are
also independent from each other in EEG data or not? In other words, is the
information of EEG windows (features) within each channel are independent
from each other?

If they are independent, so I can put them as rows too, so the number of
rows would be 10*80 (number of subjects * number of windows in each
channel).

If they are dependent, the rows are subjects only.

Thank you very much in advance for your help.

Regards,
Ali
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20180620/ea1a2c29/attachment.html>