Hi.
> Where do we find the random effect variance (REV)? Mixed effects variance
> (MEV)? averaged Fixed effects variance?
> we suspect to find these in the following files:
>
>
> *.gfeat/varcope1
This is the variances from the lower-level which feed into the
mixed-effects variance estimation. As such it can be considered the
(not-yet-pooled) fixed-effects variance; it's probably not much use to
you.
> *.gfeat/cope1.feat/stats/varcope1
this is the mixed-effects variance, ie the variance associated with the
estimate of the effect (cope1).
> *.gfeat/cope1.feat/stats/mean_random_effects_var1
this is the mean posterior random effects variance - probably not much use
to you.
> is it really the MEV that is used to generate the z-scores?
Kind of, yes. It is in the case of OLS. In the case of FLAME, the final T
distribution is estimated directly with EM/MCMC, not via cope/se(cope).
Hope this makes sense! Steve.
Stephen M. Smith
Head of Image Analysis, FMRIB
Oxford University Centre for Functional MRI of the Brain
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
|