Dear Cris,
I am working on this one... However, I just came across another serious
bug, so there is not a perfect solution yet. I will send you some code in
private mail, but please check the covariance components carefully before
you estimate the design. The code seems to leave out some of them.
Volkmar
On Wed, 13 Sep 2006, Cris Lanting wrote:
> Dear SPM list,
>
> As mentioned earlier by M. Buehler and B.Suchan there seems to be a bug in the
> covariance component specification
> when using a flexible factorial design (ANOVA) with in my case a 3x2x2 design
> (3 factors, 2 levels, 2 levels)
>
>
> Temporal non-sphericity (over voxels) : ...REML estimation
> ??? Error using ==> plus
> Matrix dimensions must agree.
>
> Error in ==> spm_reml at 87
> C = C + Q{i}*h(i);
>
> My question is if there is a solution/update to this problem?
> (other than a work-around using for example multiple regression to do a second
> level analysis...)
>
>
> With kind regards,
>
> Cris Lanting
>
>
> ----------
> Ir C.P. Lanting
> University Medical Center Groningen
> BCN Neuroimaging Center
> Antonius Deusinglaan 2, room 125
> P.O.Box 196
> 9700 AD Groningen
> The Netherlands
> phone: +31 (0)50 3637157
> [log in to unmask]
>
--
Volkmar Glauche
-
Department of Neurology [log in to unmask]
Universitaetsklinikum Freiburg Phone 49(0)761-270-5331
Breisacher Str. 64 Fax 49(0)761-270-5416
79106 Freiburg
|