[Eeglablist] ICA question

[Makoto Miyakoshi]

Dear Francesca,

For runica(), use pca option. Type this to the option box which should have
'extended', 1 already:
'pca', 10

However keep in mind that it's better to select largest GOOD 10 ICs out of
111 than reduce dimension to 10. Using 10 ICs out of 111 does NOT mean you
are using 9% of the data because the ICs are sorted by order of variance
i.e. IC1 is typically 100 times larger in variance than IC35. The largest
good 10 ICs would surely account for more than 50% of your data. Use
envtopo() function which you can find in the main GUI 'Plot'-'Component
ERPs'-'With component maps' to test percent variance accounted for (pvaf).

Makoto


2013/10/4 Francesca Miraglia <fra.miraglia at gmail.com>

> Dear all,
> I'd like to ask you an ICA question:
>
> is it possible to force the number of ICA components?
>
> I have a dataset of 111 eeg channels and I'd like that all the 111
> possible components converge only in the first 10 indipendent components.
>
> Can someone tell me how can I solve this dilemma?
>
> Thanks in advance.
>
> Best regards,
>
> Francesca Miraglia,
> PhD candidate, Rome
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Makoto Miyakoshi
Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20131004/a399d94d/attachment.html>