[FieldTrip] Clustering a conjunctive stat map

Nick Ketz nick.ketz at gmail.com
Wed Feb 15 02:40:45 CET 2017


Hi trippers,

I'm considering implementing some code to cluster a stat matrix that was
calculated using the ft_conjunctiveanalysis function.

Looking through the code and reading up on this form of conjunctive
statistics, my impression is that I could do a similar clustering of this
stat output if I hold on to the original statrand matrixes calculated in
ft_statistics_montecarlo.m for each of the stat structs that are being
input into ft_conjunctiveanalysis.  If I have those statrand mats present,
I can do a similar minimum statistic conjunction across the various stat
objects to create an analogous statrand matrix that can then be used with
the conjunction of the true data as input into clusterstat.m

I'd like to make this workable as a possible submission to the fieldtrip
code base, but didn't want to dig into it unless it was statistically
sound/made any sense to begin with.  So, please chime in if I'm missing
something and the cluster of on this conjunction of statrands is nonsense,
but also please chime in if this does make sense and I can proceed as
described.


Nick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20170214/123eef17/attachment.html>


More information about the fieldtrip mailing list