[Eeglablist] statcondfieldtrip for cluster-based nonparametric/permutation testing
thearn at asu.edu
Sun Feb 24 15:12:24 PST 2019
I am trying to use the "statcondfieldtrip" function to analyze time-series
and time/frequency event-related potentials, and I am looking for some
assistance. When I attempt to run "statcondfieldtrip" I get the following
error: Error using ft_freqstatistics (line 177) the length of the design
matrix (3) does not match the number of observations in the data (6).
My data is organized as a 2x3 cell (2 conditions and 3 groups) with each
element containing a matrix of (time x channel x subject) or (time x
frequency x channel x subject) data. The function is called like this:
[stats, df, pvals] = statcondfieldtrip(data, 'paired', 'on', 'method',
'permutation', 'naccu', '1000', 'neighbours', neighbours, 'alpha', 0.05,
The "neighbours" structure is created using the "ft_prepare_neighbours"
function (template method) with no issues.
I'm not sure if this is a syntax error on my part or a missing input. Any
assistance would be greatly appreciated!
Biomedical Engineering PhD Candidate
Arizona State University
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the eeglablist