<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=ks_c_5601-1987">
<META content="MSHTML 6.00.2900.3059" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><FONT face="¸¼Àº °íµñ" size=2>How are you?</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>Recently, I have been requested for analyzing my
ERP data using PCA.</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>Therefore, i implanted the PCA plugin
fuction(from "plugins" section in the WEB site of yours).</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2><FONT face="¸¼Àº °íµñ" size=2>But i can't explain
what kinds of the PCA algorithms was used for analysing in my
manuscript.</FONT></FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>Regarding this, I'm just wondering
what the kind of PCA algorithms (e.g. tPCA or Sequencial PCA) </FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>were conducted in the EEGLAB
process. </FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>How can i expalin about the PCA algorithm in
EEGLAB at the Manuscrpit?</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2></FONT> </DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>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. </FONT><FONT face="¸¼Àº °íµñ" size=2>In the
case of ICA, I can find the ppaf value of each component.
</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>Synthax below is the solution which i've found
from mailing list.</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>However, the eigenvlue which is resulted from
synthax has too high value (e.g. 4403.50). </FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>Plz tell me exatly how can i get the eigenvalue
from below synthax.</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2></FONT> </DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>------</FONT></DIV>
<DIV><FONT face=Arial size=2>[pc,eigvec,sv] = runpca(EEG.data(:,:));<BR><BR>%
Outputs:<BR>% pc - the principal components, i.e. >> inv(eigvec)*data =
pc;<BR>% eigvec - the inverse weight matrix (=eigenvectors). >> data =
eigvec*pc;<BR>% sv - the singular values (=eigenvalues)<BR><BR>You may obtain
the percentage variance accounted for by using the <BR>compvar() function (enter
PCA components instead of ICA components), <BR>i.e. to get the percentage
accounted for by the first PCA component<BR><BR>[proj pvaf] =
compvar(EEG.data(:,:), pc, eigvec, 1);<BR>% pvaf = percentage of variance
accounted for<BR>-------------</FONT><BR></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>Surely, i go through the mailing
list. </FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>The a</FONT><FONT face="¸¼Àº °íµñ" size=2>rchieved
list, however, is too difficult to understand for me, as a very low level
user.</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>Anyone can explain the PCA function in
EEGLAB?</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>or </FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>Anyone can tell the method to carry out PCA for
EEGLAB matfile(EEG.data)?</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2></FONT> </DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>Plz help me</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2></FONT> </DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>+</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>Tae-Ho Lee</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>Lab. of Behavioral Neuroscience.</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2>Psychology of Korea Univ.</FONT></DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2></FONT> </DIV>
<DIV><FONT face="¸¼Àº °íµñ" size=2></FONT> </DIV></BODY></HTML>