r-boun...@nmr.mgh.harvard.edu
On Behalf Of Anderson M. Winkler
Sent: Wednesday, July 31, 2019 11:10 AM
Cc: freesurfer@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] Running FSL randomise on FS cortical thickness data
{Disarmed}
External Email - Use Caution
Hi,
It's possible to analyse longitudi
External Email - Use Caution
Hi,
It's possible to analyse longitudinal designs with PALM. Some mild
assumptions must hold, and those assumptions are different than those of an
LME model. PALM doesn't assume normality, doesn't require a specified form
for the dependence structure,
I'm not sure that is a way to do this as it is quite tricky managing the
exchangeability in longitudinal data. Maybe PALM can do it. I'm cc'ing
Anderson in case he wants to weigh in.
On 7/28/19 7:24 PM, Mark Wagshul wrote:
>
> External Email - Use Caution
>
> Doug,
>
> Thanks. Unfortuna
External Email - Use Caution
Doug,
Thanks. Unfortunately, this is only for the GLM, is there an equivalent
version for the linear mixed model? I haven’t seen one. I guess we could
figure out how to integrate lme into mri_glmsim, but I’m not sure that’s a
better use of our tim