[FieldTrip] [EXTERNAL] Re: New issue with ft_statistics_montecarlo

Matthew Wood matthew.wood at utsa.edu
Fri May 28 21:19:28 CEST 2021


Thank you for your help.

My cfg.clusterthreshold was set to the default "parametric". I changed it to "nonparametric_common" and now ft_freqstatistics runs without any errors.

Thank you!
From: fieldtrip <fieldtrip-bounces at science.ru.nl> on behalf of Schoffelen, J.M. (Jan Mathijs) via fieldtrip <fieldtrip at science.ru.nl>
Sent: Friday, May 28, 2021 1:54 PM
To: FieldTrip discussion list <fieldtrip at science.ru.nl>
Cc: Schoffelen, J.M. (Jan Mathijs) <jan.schoffelen at donders.ru.nl>
Subject: [EXTERNAL] Re: [FieldTrip] New issue with ft_statistics_montecarlo

  This email originated outside of The University of Texas at San Antonio.
  Please exercise caution when clicking on links or opening attachments.

Hi Matthew,

The error tells you that it is impossible to determine a parametric critical value for clustering.

You should either:

-specify a cfg.statistic that points to an ft_statfun_something that internally contains some code that does the computation of the critical value. This typically pertains to a statistic for which known parametric distributions exist, e.g. a T-statistic, or an F-statistic.

-explicitly set a cfg.clustercritval to be an upper and lower numeric threshold for clustering.

-use a different specification of cfg.clusterthreshold, e.g. ’nonparametric_common’.

Best wishes,

On 28 May 2021, at 19:45, Matthew Wood via fieldtrip <fieldtrip at science.ru.nl<mailto:fieldtrip at science.ru.nl>> wrote:


I have had a script for my statistics that uses ft_statistics_montecarlo for the method in ft_freqstatistics. I downloaded the latest version of FieldTrip recently and now my scripts do not work. I get the following errors:

Error using ft_statistics_montecarlo (line 250)
could not determine the parametric critical value for clustering

Error in ft_freqstatistics (line 194)
 [stat, cfg] = statmethod(cfg, dat, design);

I'm not sure what could be driving this issue, especially since I have made no changes to the script. The only thing that changed was downloading the lastest version of Fieldtrip. Is anyone else experiencing this? Or does anyone have advice into how it works?

I've looked up the error message and one user recommended testing that running with cfg.method = 'analytic' works. I've done this and my script runs fine. The structure it outputs gives me appropriate values (no NaNs).

I'd be happy to give more context if needed.

Thank you,
fieldtrip mailing list

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20210528/e6b6470e/attachment-0001.htm>

More information about the fieldtrip mailing list