Back to the main page.

Bug 2813 - ft_statfun_diff_itc doesn't create/output a critval

Status CLOSED INVALID
Reported 2015-01-23 11:44:00 +0100
Modified 2019-08-10 12:30:45 +0200
Product: FieldTrip
Component: core
Version: unspecified
Hardware: PC
Operating System: Windows
Importance: P5 enhancement
Assigned to:
URL:
Tags:
Depends on:
Blocks:
See also:

Johanna - 2015-01-23 11:44:36 +0100

Hi all, Firstly, thank you for the great statfun of 'diff_itc'. I tried using it with cfg.method='montecarlo' and cfg.correctm='cluster'. However, this crashes due to line 251 of ft_statistics_montecarlo not being able to get the 'critval' from this statfun. Am I understanding this correctly that it should be able to be used this way? If so, would be great to compute this extra feature (though I'm not sure how to compute the critval for this). Alternatively, should only other cfg.correctm values (other than 'cluster') be used with this method? If so, then I can update the documentation of 'diff_itc' so that it makes this clear. Thank you, Johanna


Jan-Mathijs Schoffelen - 2015-01-23 11:54:20 +0100

I am not aware of a parametric thresholding scheme that quantifies a critical value, but it may be in line with how a critical value for coherence difference can be determined (doing something with atanh() and a dof estimation of 1./sqrt(2*dof), see Maris, Schoffelen et al :-). You could still use clustering, but then specifying cfg.clusterthreshold to be either 'nonparametric_common' or 'nonparametric_individual'


Robert Oostenveld - 2015-01-23 12:01:53 +0100

(In reply to Jan-Mathijs Schoffelen from comment #1) I agree with JM: . I will add something along the line of For this specific statistic there is no known parametric distribution, hence the probability and critical value cannot be computed. This specific statistic can therefore only be used with cfg.method=montecarlo. If you want to do this in combination with cfg.correctm=cluster, you need cfg.clusterthresholdnonparametric_common or cfg.clusterthreshold=nonparametric_individual. to the documentation of the function.


Robert Oostenveld - 2015-01-23 12:04:29 +0100

roboos@mentat001> svn commit ft_statfun_diff_itc.m Sending ft_statfun_diff_itc.m Transmitting file data . Committed revision 10118.


Robert Oostenveld - 2019-08-10 12:30:45 +0200

This closes a whole series of bugs that have been resolved (either FIXED/WONTFIX/INVALID) for quite some time. If you disagree, please file a new issue on https://github.com/fieldtrip/fieldtrip/issues.