[Eeglablist] how to apply ICA repeatedly to a data set

[arno]

Shivaprakash wrote:
> I have a doubt regarding how to apply ICA to a data set repeatedly. I 
> want to do that _from the matlab command-line_. Is the components so 
> obtained after repeated application of ICA better than those obtained 
> by applying ICA just once?
No. They are the same. If you used the first decomposition to remove 
artefactual epochs, the second decomposition might be of better quality.
> what is the measure to find out the amount of improvement after say 
> from  ICA applied once to the second application of ICA to the same 
> data. I want the measure of improvement to decide when to stop.
There is no perfect measure. One measure is their independence using 
mutual information for instance. Another measure is their dipolar 
quality (how well can they be modeled using a dipole model).

Arno
>  
> thanks in advance
>  
> G. Shivaprakash.
>
> ------------------------------------------------------------------------
> Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great 
> rates starting at 1¢/min. 
> <http://us.rd.yahoo.com/mail_us/taglines/postman7/*http://us.rd.yahoo.com/evt=39666/*http://beta.messenger.yahoo.com> 
>
>
>  <http://us.rd.yahoo.com/mail_us/taglines/postman7/*http://us.rd.yahoo.com/evt=39666/*http://beta.messenger.yahoo.com>
> ------------------------------------------------------------------------
>
> _______________________________________________
> eeglablist mailing list eeglablist at sccn.ucsd.edu
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu <http://us.rd.yahoo.com/mail_us/taglines/postman7/*http://us.rd.yahoo.com/evt=39666/*http://beta.messenger.yahoo.com>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20060420/c1ec9554/attachment.html>