[Eeglablist] Bad channel removal

[mori larin]

Dear all,

I have a question about bad channel removal in EEGLAB using kurtosis
greater than 5.
The threshold (5) is highly arbitrary. Also, kurtosis is sensitive to
outliers and few bad values can easily allow this threshold to be crossed.
So why the kurtosis has been used with specific threshold?

Many thanks,
Mori
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20150317/be2d1ac8/attachment.html>