<html>
<head>
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">Hi,<br>
<br>
thanks a lot for your input. I tried Simons PCA approach on a
couple of datasets and it seems like I need less than half of my
components to explain 99% of variance in my high high frequency
data. <br>
<br>
I am wondering, though, if this approach is appropriate since PCA
and ICA work differently. As far as I understood, PCA tries to
maximize the variance explained of each component, whereas ICA
tries to maximize the independence of the components. So if, for
instance, 64 PCs explain 99% of my variance, 64 ICs might explain
much less.<br>
<br>
Another thing that seems worth mentioning is that when I run the
same procedure over the low pass data or the original unsplit data
I need even fewer PCs to explain the same percentage of variance.
So this procedure does not really explain why my ICA on the high
pass filtered data takes so much longer than the ICAs in low
frequency or unsplit data. <br>
<br>
Could it be that it's a question of independence rather than of
variance explained? Is there a way to estimate how many
independent sources there are in the data?<br>
<br>
Thanks!<br>
<br>
Martin<br>
<br>
<br>
<br>
On 17.05.2013 00:25, Simon-Shlomo Poil wrote:<br>
</div>
<blockquote
cite="mid:CAKe6oahwMLAbFm=Wf-VwUABfyX=aTwS=7bvuaEg7JtdxZWvRFQ@mail.gmail.com"
type="cite">
<meta http-equiv="Content-Type" content="text/html;
charset=ISO-8859-1">
<meta name="Generator" content="MS Exchange Server version
08.03.0279.000">
<title>Re: [Eeglablist] ICA on lowpass / highpass filtered data</title>
<!-- Converted from text/plain format -->
<p><font size="2">Dear Martin,<br>
<br>
As Makoto says, you make the channels less independent of each
other.<br>
It might be resonable* to reduce using PCA. One way to
determine the<br>
number of relevant dimesions could be<br>
[COEFF, SCORE, LATENT] = princomp(EEG.data');<br>
tmp = cumsum(LATENT);<br>
nr=find(tmp/tmp(end)>0.975,1);<br>
<br>
, which gives you the number of principle components
explaining 97.5 %<br>
of the variance.<br>
<br>
*you can find previous mails discussing pro-/con- of PCA
reduction on<br>
the this list (I remember there was even a paper in prep? I
didn't see<br>
it come out)<br>
<br>
Best wishes<br>
Simon<br>
<br>
--<br>
Simon-Shlomo Poil, Dr.<br>
<br>
2013/5/16 Makoto Miyakoshi <a class="moz-txt-link-rfc2396E" href="mailto:mmiyakoshi@ucsd.edu"><mmiyakoshi@ucsd.edu></a>:<br>
> Dear Martin,<br>
><br>
> If you apply a band-pass filter, your channel data become
less independent<br>
> of each other i.e. rank-reduced.<br>
><br>
> Imagine you apply an extreme band-pass filter, say
10-11Hz. All of your<br>
> channel data look very much like each other.<br>
><br>
> Makoto<br>
><br>
><br>
> 2013/5/16 Krebber, Martin
<a class="moz-txt-link-rfc2396E" href="mailto:martin.krebber@charite.de"><martin.krebber@charite.de></a><br>
>><br>
>> Hi all,<br>
>><br>
>><br>
>> I am currently working on an analysis were I split
the data into low and<br>
>> high frequency portions using a lowpass (cutoff 35
Hz) and a highpass<br>
>> (20 Hz) filter, respectively. The idea behind this
approach is to do the<br>
>> ICA artefact rejection seperately on low and high
frequency data in<br>
>> order to be better able to reject high frequency
muscle artefacts and<br>
>> obtain a clearer brain signal in the gamma range.<br>
>><br>
>> My problem is that, especially with the highpass
filtered data, ICA<br>
>> takes a very long time (roughly 5-10 times the usual)
and even then the<br>
>> decomposition does not look very clean. I tried to
reduce the<br>
>> dimensionality of the data (from 128 to 96) by
applying the PCA<br>
>> parameter in pop_runica and it is way faster. Is it
justified, or maybe<br>
>> even recommended to reduce the data dimensionality
after filtering out a<br>
>> considerable portion of the signal? And if so, is
there a rule of thumb<br>
>> about how much to reduce the data dimensionality?<br>
>><br>
>> Thanks for any suggestions!<br>
>><br>
>> Regards,<br>
>> Martin<br>
>><br>
>> _______________________________________________<br>
>> Eeglablist page: <a moz-do-not-send="true"
href="http://sccn.ucsd.edu/eeglab/eeglabmail.html">http://sccn.ucsd.edu/eeglab/eeglabmail.html</a><br>
>> To unsubscribe, send an empty email to<br>
>> <a class="moz-txt-link-abbreviated" href="mailto:eeglablist-unsubscribe@sccn.ucsd.edu">eeglablist-unsubscribe@sccn.ucsd.edu</a><br>
>> For digest mode, send an email with the subject "set
digest mime" to<br>
>> <a class="moz-txt-link-abbreviated" href="mailto:eeglablist-request@sccn.ucsd.edu">eeglablist-request@sccn.ucsd.edu</a><br>
><br>
><br>
><br>
><br>
> --<br>
> Makoto Miyakoshi<br>
> Swartz Center for Computational Neuroscience<br>
> Institute for Neural Computation, University of
California San Diego<br>
<br>
<br>
<br>
--<br>
</font>
</p>
</blockquote>
<br>
</body>
</html>