[Eeglablist] [How can i] Q abour PCA in the EEGLAB.

[Tae-Ho Lee]

How are you?
Recently, I have been requested for analyzing my ERP data using PCA.
Therefore, i implanted the PCA plugin fuction(from "plugins" section in the WEB site of yours).
But i can't explain what kinds of the PCA algorithms was used for analysing in my manuscript.
Regarding this,  I'm just wondering what the kind of PCA algorithms (e.g. tPCA or Sequencial PCA) 
were conducted in the EEGLAB process. 
How can i expalin about the PCA algorithm in EEGLAB at the Manuscrpit?

Another question is about the eigenvalue. In my knowledge, there have to be the eigenvalue of the component in the results, however, i can't find the value at al. In the case of ICA, I can find the ppaf value of each component. 
Synthax below is the solution which i've found from mailing list.
However, the eigenvlue which is resulted from synthax has too high value (e.g. 4403.50). 
Plz tell me exatly how can i get the eigenvalue from below synthax.

------
[pc,eigvec,sv] = runpca(EEG.data(:,:));

% Outputs:
% pc - the principal components, i.e. >> inv(eigvec)*data = pc;
% eigvec - the inverse weight matrix (=eigenvectors). >> data = eigvec*pc;
% sv - the singular values (=eigenvalues)

You may obtain the percentage variance accounted for by using the 
compvar() function (enter PCA components instead of ICA components), 
i.e. to get the percentage accounted for by the first PCA component

[proj pvaf] = compvar(EEG.data(:,:), pc, eigvec, 1);
% pvaf = percentage of variance accounted for
-------------

Surely, i go through the mailing list. 
The archieved list, however, is too difficult to understand for me, as a very low level user.
Anyone can explain the PCA function in EEGLAB?
or 
Anyone can tell the method to carry out PCA for EEGLAB matfile(EEG.data)?

Plz help me

+
Tae-Ho Lee
Lab. of Behavioral Neuroscience.
Psychology of Korea Univ.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20070314/0f5d8bea/attachment.html>