[Eeglablist] hi

stephan grimault stgrimault at yahoo.fr
Tue Dec 6 06:32:42 PST 2005


hi

first : Thanks for this toolbox!!!

I tried to used crossf to determine the
crosscorrelation between 2 electrodes. I edit the help
crossf but would like more informations :

first :

crossf() - Returns estimates and plots event-related
coherence (ERCOH) between two input data time series. 

How you compute the coherence? Do you have some
references to explain the calculus? Do you compute the
phase and amplitude of the coorelation function or an
another way?

A lower panel (optionally) shows the coherence phase
difference between the processes. In this panel:
-90 degrees (blue)   means x leads y by a quarter
cycle.
90 degrees (orange) means y leads x by a quarter
cycle.
 
Second :
How you know that X leads Y ?  How you unwrap the
phase difference if this quantity is grater than 2*pi?

third : 
there is the option freqmax. But why does not exist
the option freqmin?

thank by advance

Stephan GRIMAULT

CNRS, IFR 131 Science du cerveau et de la cognition,
centre IRMF Marseille France.



	
	
		
___________________________________________________________________________
Appel audio GRATUIT partout dans le monde avec le nouveau Yahoo! Messenger 
Téléchargez cette version sur http://fr.messenger.yahoo.com



More information about the eeglablist mailing list