[FieldTrip] Why the general use of parametric test as observed test statistic for permutation cluster tests?

Peter Goodin pgoodin at swin.edu.au
Wed Apr 15 02:06:33 CEST 2015

Hi Fieldtrip list, 

I'm just wondering if someone can explain if there is a reasoning behind the general use of a parametric test to calculate the initial step in the permutation cluster test and not a non-parametric test (e.g t-test over Mann Whitey U)? Is it just due to the assumption of a Gaussian distribution or is there something deeper going on? 



Peter Goodin, 
BSc (Hons), Ph.D Candidate (submitted).

Brain and Psychological Sciences Research Centre (BPsych)
Swinburne University, 
Hawthorn, Vic, 3122

Monash Alfred Psychiatry Research Centre (MAPrc)
Level 4, 607 St Kilda Road, 
Melbourne 3004

More information about the fieldtrip mailing list