In theory it is because the F can be converted to a p which can be converted into a pseudo-z. When you run mri_glmfit-sim, you should make sure to use an unsigned test ("abs")
On 12/22/17 8:08 PM, basile pinsard wrote:
Hi Freesurfer experts. Does mri_glmfit-sim Monte-Carlo cluster size correction is also usable for F-test? The doc says that it synthesizes a z map (I assume from a normal distribution) for computing the voxel-wise thresholds and cluster size distribution. Maybe I am wrong with my stats but I wonder if it is ok to use this null-hypothesis to a F-distributed data such as results from a F-test. Thanks!
-- Basile Pinsard /PhD candidate, / Laboratoire d'Imagerie Biomédicale, UMR S 1146 / UMR 7371, Sorbonne Universités, UPMC, INSERM, CNRS /Brain-Cognition-Behaviour Doctoral School /*, *ED3C*, *UPMC, Sorbonne Universités Biomedical Sciences Doctoral School, Faculty of Medicine, Université de Montréal CRIUGM, Université de Montréal
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer