[Eeglablist] changing the reference electrode for use the SIFT toolbox

[Tim Mullen]

Thanks John! Glad you find it useful.

answers below:

i) This line replaces a point estimate of the connectivity estimator
(computed from all available trials) by the expected value of the estimator
obtained from the bootstrap distribution. You can think of the bootstrap
distribution (PConn_boot) as a "plugin" for the population distribution of
the connectivity estimator. Under mild assumptions, this converges to the
population distribution as the number of bootstrap "iterations" (samples)
tends to infinity. So the mean, variance, quantiles, and other statistics
of this distribution are the quantities of interest. That being said, you
could also just use the original point estimate. This represents a
particular sample from the bootstrap distribution -- the (very low
probability) case where you happened to draw each trial exactly once when
sampling with replacement to construct the surrogate dataset for a given
bootstrap iteration.

ii) The non-parametric (surrogate) statistical routines are equally valid
for all of the connectivity measures. The parametric (analytic) tests are
applicable for PDC, nPDC, RPDC.

Best,
Tim


On Wed, Nov 13, 2013 at 12:23 PM, jfochoaster . <jfochoaster at gmail.com>wrote:

> Thanks Dr. Tim,
>
> I am using your toolbox for around one year, is a great toolbox
>
> I have two doubt in the ScriptingExample,
>
> i) Why do you use the following instruction?
>
> EEG(cnd).CAT.Conn = stat_getDistribMean(PConn_boot(cnd));
>
> ii) For which measures are tested the statistical routines?
>
> Thanks in advance
>
>
> On Wed, Nov 13, 2013 at 3:12 PM, Tim Mullen <mullen.tim at gmail.com> wrote:
>
>> Dear John,
>>
>> If you are using connectivity measures with channel data, an average
>> reference or laplacian reference are better choices than linked mastoid
>> reference.  See the following paper for a good discussion in the context of
>> coherence analysis (with applicability also to Granger causality):
>>
>> Nunez, P. L., Srinivasan, R., Westdorp, A. F., Wijesinghe, R. S., Tucker,
>> D. M., Silberstein, R. B., & Cadusch, P. J. (1997). EEG coherency. I:
>> Statistics, reference electrode, volume conduction, Laplacians, cortical
>> imaging, and interpretation at multiple scales. *Electroencephalography
>> and clinical neurophysiology*, *103*(5), 499–515.
>> Be aware that connectivity measures applied to channels will generally
>> suffer from an increased rate of spurious coupling (false positives) due to
>> volume conduction. Spatial laplacians can help mitigate this.
>>
>> If you are using components then the choice of reference is irrelevant
>> (ICA produces a reference-free signal).
>>
>> Tim
>>
>>
>>  On Fri, Nov 8, 2013 at 2:20 PM, jfochoaster . <jfochoaster at gmail.com>wrote:
>>
>>>  Dear EEG experts,
>>>
>>> I have a lot of eeg data with the average as reference. This data comes
>>> originally with one electrode near to CZ as reference
>>>
>>> Is valid re-reference the data to linked mastoid?
>>>
>>> This is because I am using the SIFT toolbox over the channels and the
>>> DFT measure is applied over eeg data with linked mastoid as reference
>>>
>>> If I use components instead of channels, I can forget the reference?
>>>
>>> Thanks in advance
>>>
>>> --
>>> John Ochoa
>>> Docente de Bioingeniería
>>> Universidad de Antioquia
>>>
>>> _______________________________________________
>>> 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
>>>
>>
>>
>>
>> --
>> ---------  αντίληψη -----------
>>
>
>
>
> --
> John Ochoa
> Docente de Bioingeniería
> Universidad de Antioquia
>



-- 
---------  αντίληψη -----------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20131113/d1ce9456/attachment.html>