[FieldTrip] Number of randomizations + 1 in clustering tests?
"Jörn M. Horschig"
jm.horschig at donders.ru.nl
Wed Mar 5 14:58:28 CET 2014
Hi Mariana,
yes, this is the case. You can think of it as your original data also
being an observation, thus needs to be added to the number of
permutations. If I remember correctly, this is for example explained
here in more detail (and probably in more correct terms):
http://www.statsci.org/smyth/pubs/permp.pdf
Best,
Jörn
On 3/5/2014 2:37 PM, Mariana Rebelo wrote:
> Dear all
>
> I am computing some clustering tests on my MEG data and I don't
> understand the p-value returned in the field stat.posclusters(1).prob
> If I perform 1000 randomizations I get a p-value of 0.00799201
> I don't understand how this is possible, the p-value should be equal
> to integer/1000, so I should get either 0.0070 or 0.0080
> It seems that in fact I am not performing 1000 randomizations as I set
> in numrandomization, but 1001, because then: 8/1001 = 0.00799201
>
> Is this really the case? Is the fieldtrip script adding another
> randomization to what is set by the user?
>
> Thanks!
> Mariana
>
>
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
--
Jörn M. Horschig
PhD Student
Donders Institute for Brain, Cognition and Behaviour
Centre for Cognitive Neuroimaging
Radboud University Nijmegen
Neuronal Oscillations Group
FieldTrip Development Team
P.O. Box 9101
NL-6500 HB Nijmegen
The Netherlands
Contact:
E-Mail: jm.horschig at donders.ru.nl
Tel: +31-(0)24-36-68493
Web: http://www.ru.nl/donders
Visiting address:
Trigon, room 2.30
Kapittelweg 29
NL-6525 EN Nijmegen
The Netherlands
More information about the fieldtrip
mailing list