[Eeglablist] Fwd: SIFT analytic statistics

[Octavian Lie]

Any help with this?
Thank you,

Octavian


---------- Forwarded message ----------
From: Octavian Lie <octavian.lie at gmail.com>
Date: Mon, May 18, 2015 at 11:07 AM
Subject: SIFT analytic statistics
To: "eeglablist at sccn.ucsd.edu" <eeglablist at sccn.ucsd.edu>


Dear All,

I am a new member. I tried to use the analytic statistics code for
significance for the DTF measures based on Eichler 2006, and it looks that
this is not available in the last (beta) release. Any plans to implement
that soon?

On the same note, I am using adaptive DTF measures on nonstationary time
series (seizures). I am not sure the pointwise significance levels
calculated as in Eichler would accommodate the time varying case. One
option is to break down nonstationary EEG time series I work
with in several epochs of relative stationarity, and calculate pointwise
significance on these epochs. Is there a way to use *est_checkMVARStability()
*to accomplish this? I mention I use a Kalman based TVAR process and not
the window based approach implemented in SIFT. If not, any alternatives
suggested?

Thank you,

Octavian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20150520/e12be9af/attachment.html>