Dear Guillaume, thank you for your answer, it helps a lot.
After looking into these lines, I've found that (spm_fmri_spm_ui.m, line337-348) if xVi.V is not set as i.i.d., the AR(0.2) was assumed in structure of xVi.Vi .
I am curious about:
Q1. Why always AR(0.2) ? Shouldn't we use exp(-1) in exponential correlation model ? (as indicated in SPM book (2007))
Q2. I tried to understand the detail of generating xVi.Vi and I looked in to spm_Ce.m
% dCda
%----------------------------------------------------------------------
if ~isempty(a)
Q = spm_Q(a,v);
dQda = spm_diff('spm_Q',a,v,1);
C{1} = Q - dQda{1}*a;
C{2} = Q + dQda{1}*a;
else
C{1} = speye(v,v);
end
I can understand spm_Q is the exponential correlation, but I don't understand the remaining lines. Why should xVi.Vi be generated in this way ?
Thanks for your time, very appreciate for your help.
Sincerely
Philip Lin
On Fri, 20 Dec 2013 11:44:39 +0000, Guillaume Flandin <[log in to unmask]> wrote:
>Dear Philip,
>
>have a look at spm_reml.m lines 194-197 and spm_spm.m line 880.
>You can recompute V (=SPM.xVi.V) with:
>
>V = 0;
>for i = 1:numel(SPM.xVi.Vi)
> V = V + SPM.xVi.Vi{i} * SPM.xVi.h(i);
>end
>V = V * SPM.nscan/trace(V);
>
>Best wishes,
>Guillaume.
>
>
>On 20/12/13 04:34, Philip Lin wrote:
>> Hi all,
>>
>> I've run the "Auditory" example in SPM8 manual step by step (ch28, p209-222) and met some problems from the output.
>>
>> Q1. As I know, the covariance structure V was approximated by a linear formula: a1*Q1 + a2*Q2, where Q1 is identity matrix and Q2 is an AR(1) exponential correlation matrix. The hyperparameters (a1, and a2) were estimated by ReML. Thus, the estimated V could be calculated by hand (because a1, and a2 were given in SPM.xVi.h), but why the result in this way is different to SPM.xVi.V ? And there're two cells in SPM.xVi.Vi, what's the difference ? I've checked the spm_spm.m (line 29-43), but I still can't get it.
>>
>> many thanks
>>
>> Philip Lin
>>
>
>--
>Guillaume Flandin, PhD
>Wellcome Trust Centre for Neuroimaging
>University College London
>12 Queen Square
>London WC1N 3BG
|