Hi Mina,
a simple matlab trick that can be used just for visualization purposes
(but only of A, B and C values, as pointed out by Mohamed) could be the
following (it is not actually a script, but it can be easily adapted):
-------------------------------------------------
clear
load DCM_subject01; % DCM model of one of your subjects included in the
averaging
M=DCM.M;
U=DCM.U;
Y=DCM.Y;
xY=DCM.xY;
H1=DCM.H1;
H2=DCM.H2;
K1=DCM.K1;
K2=DCM.K2;
R=DCM.R;
y=DCM.y;
T=DCM.T;
Ce=DCM.Ce;
F=DCM.F;
clear DCM;
load DCM_avg_model; % your averaged DCM model
DCM.M=M;
DCM.U=U;
DCM.Y=Y;
DCM.xY=xY;
DCM.H1=H1;
DCM.H2=H2;
DCM.K1=K1;
DCM.K2=K2;
DCM.R=R;
DCM.y=y;
DCM.T=T;
DCM.Ce=Ce;
DCM.F=F;
save DCM_avg_model_visualization; % to be used just for visualizing A,B
and C values
clear
-------------------------------------------------
Best,
Marco
Mohamed Seghier wrote:
>
> Hi Mina,
>
> In the current version of SPM5, I think it is not possible to use the
> REVIEW option to explore an average DCM model. When averaging your DCM
> models, the structure DCM.M, needed for instance in the REVIEW option,
> is not available (i.e. not updated in spm_dcm_average).
>
> For the average option, you need first to estimate your individual
> models (one per run). The AVERAGE option will then estimate the
> "average" of each parameter weighted by its precision (for matrices A, B
> and C only).
> To display your average DCM model, load it in Matlab and have a look to
> the parameter matrices (A, B and C) and their posterior probability
> matrices (pA, pB, pC)..
>
> See also Klaas's reply in the same topic:
> http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind06&L=spm&O=D&F=&S=&P=242866
>
>
> Good luck,
>
> Mohamed
>
>
>
>
> Mina Khoshnejad wrote:
>> Hello,
>>
>> I have a question about the *Average Option* in DCM. When I average up
>> my different DCM.mat corresponds to different runs, I am not able to
>> see the results like before (*Using the Review Option*).
>>
>> It does not *Estimate* the new created Average.DCM.mat, and does not
>> display it with Review. Anyone has any idea how it works?
>>
>> Many Thanks,
>> Mina
>>
>>
>> between 0000-00-00 and 9999-99-99
>>
>
--
Marco Tettamanti, Ph.D.
San Raffaele Scientific Institute
Facoltà di Psicologia
Via Olgettina 58
I-20132 Milano, Italy
Tel. ++39-02-26434888
Fax ++39-02-26434892
Email: [log in to unmask]
|