Javier,
specifying equal variance in each factor is probably an invalid assumption,
but you should carefully decide which factors will have unequal variance
across their levels.
Volkmar
On Wed, 18 Oct 2006, Javier Sánchez wrote:
> You are right. I have to state equal variance in each factor and it works.
>
> Thank you very much for your help.
>
> Javier
>
> > On Wed, 18 Oct 2006, Javier Sánchez wrote:
> >
> > > Dear Volkmar,
> > > thank you very much for your help. I am
> > trying to run the SPM using your *.m
> > > files but I have another problem. Now after
> > model estimation SPM changes the
> > > desing matrix to complex values with some imaginary elements.
> > > I can not even use the review button of SPM to check the desing
> > > matrix
> >
> > You should review the design _before_ estimation, not after... I suspect
> > that you may have specified all factors having unequal and dependent
> > variances. This will not work, as the trick with statistics is to make the
> > proper assumptions about which variances can be treated as equal and
> > independent.
> >
> > >
> > > Can you figure out what might be the problem?
> > >
> > > Thanks again,
> > >
> > > Javier
> > > > Javier,
> > > >
> > > > if you feel adventurous, you could try patching SPM5 with these
> > > > 2 files
> > > > instead of the original ones.
> > > > Before you estimate, you should check whether the error
> > > > covariances will
> > > > be modeled as you want them. Use
> > > > ReviewDesign->Explore->Covariance and
> > > > browse through the specified component matrices.
> > > >
> > > > Volkmar
> > > >
> > > > On Wed, 18 Oct 2006, Javier wrote:
> > > >
> > > > > Hi All,
> > > > >
> > > > > I am using a Flexible Factorial model in SPM5 with three
> > > > > factors as
> > > > > shown
> > > > > in the figure attached. When I try to estimate the result
> > > > > of the
> > > > > desing
> > > > > matrix SPM produces this error.
> > > > >
> > > > > Temporal non-sphericity (over voxels) :
> > > > > ...REML
> > > > > estimation
> > > > > ??? Error using ==> *
> > > > > Inner matrix dimensions must agree.
> > > > >
> > > > > Error in ==> c:\spm5\spm_reml.m
> > > > > On line 95 ==> iCX = iC*X;
> > > > >
> > > > > Error in ==> c:\spm5\spm_spm.m
> > > > > On line 858 ==> [V,h] = spm_reml(Cy,xX.X,xVi.Vi);
> > > > >
> > > > > Error in ==> c:\spm5\spm_getSPM.m
> > > > > On line 210 ==> [SPM] = spm_spm(SPM);
> > > > >
> > > > > Error in ==> c:\spm5\spm_results_ui.m
> > > > > On line 264 ==> [SPM,xSPM] = spm_getSPM;
> > > > >
> > > > > ??? Error while evaluating uicontrol Callback.
> > > > >
> > > > >
> > > > > I don't know if I am doing something wrong or if it is a
> > > > > bug of the
> > > > > SPM5
> > > > >
> > > > > Thanks in advance,
> > > > >
> > > > > Javier
> > > > >
> > > > >
> > > >
> > > > --
> > > > 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
> > > > http://fbi.uniklinik-freiburg.de/
> > > >
> > >
> > >
> >
> > --
> > 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 http://fbi.uniklinik-freiburg.de/
>
>
>
--
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 http://fbi.uniklinik-freiburg.de/
|