[Eeglablist] newtimef issues

[Arnaud Delorme]

Dear Ido,

> 1. winsize - I know that the default is frames/8. However, since I`m interested in the range of 1-200 Hz and I`d like to have good time resolution, I`m using a window of 64msec and padratio of 16. Am I going to be criticized for using such a short window?
> 

I do not think so. You need to make sure your sampling rate is high enough though (minimum 400 Hz but I would probably recommend at least 1000 Hz).

> 2. nfreqs & timesout - how does newtimef set the number of output frequencies and times? Is it a simple averaging across times and frequencies? If I`d like to downsample my data (i.e. reduce the time and frequency resolution) - is that the right way to achieve that?
> 

Output time points can be a downsampled version of the data time points. Simply look at the timefreq help message. For instance -2 will downsample twice. For frequencies you may enter the exact frequencies you are interested in.

> 3. Single trial data (tfdata) - I know that this issue has been raised a few times. I`m having some difficulties with the baseline correction of single trials timef data. I`m using this data for classification, so I don`t want to use a global baseline, which introduces dependencies between trials. On the other hand - the single trial baseline is quite noisy. Any ideas?
> 

These is no obvious solution to that problem I think. Look also at BCILAB, a new toolbox built on top of EEGLAB, that has powerful classifiers and feature extractors.

> 4. How can I change the overlap between successive windows?
> 

This is fixed by your output time points and your window size. If you change the window size and not the number of output time point, the overlap will be higher.
Thanks,

Arno


> Thanks very much for your help!
> Ido Davidesco
> 
> _______________________________________________
> 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

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20110816/637d8246/attachment.html>