Dear Dima,
SPM8b has a new interface for group BMS (access through Batch > SPM >
Stats > Bayesian Model Selection > BMS: DCM).
This new batch editor window will also appear when you click BMS at
the end of your DCM analysis.
This interface allows one to select the data (log-model evidence
values) for each subject/session/model and choose which
inference method to use (Fixed-effects or Random-effects, as described
in Stephan et. al. 2009).
If you choose Random effects then SPM will compute, among other
things, the exceedance probabilities for each model. This information
will
be displayed in the graphics window and stored in the BMS.mat file.
This file contains:
BMS.DCM.ffx (Fixed-effects analysis):
'.F' is the Free Energy (or log-model evidence) for each model/subject
'.P' is the Posterior Probability for each model (as plotted)
'.SF' contains the sum of the log-evidence (F) for each model
'.data' contains the path to the DCM files (for each subject/session
and model) that were used in this analysis.
BMS.DCM.rfx (Random-effects analysis):
'.F' is the Free Energy (or log-model evidence) for each model/subject
'.SF' contains the sum of the log-evidence (F) for each model
'.data' contains the path to the DCM files (for each subject/session
and model) that were used in this analysis.
'.exp_r' is the expected posterior probability for each model
'.xp' is the exceedance probability for each model
'.alpha' are the estimated parameters of the Dirichlet distribution
for each model
I hope this helps.
Best Regards,
Maria Joao
On Tue, Mar 31, 2009 at 2:10 PM, <Danai> <Dima>
<[log in to unmask]> wrote:
> Dear all,
>
> I just read the article 'Bayesian model selection for group studies'
>
> http://www.ncbi.nlm.nih.gov/pubmed/19306932?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_DefaultReportPanel.Pubmed_RVDocSum
>
> and I wanted to ask how can I compute the exceedance probability in ERP data. Is there already a script implemented in the SPM8b? I have already compared 2 models and I have the conditional model probability.
> I looked up the SPM_BMS, but I do not know how to run it for my EEG data. Or is there another solution?
>
> Thanks for any help
>
> Best,
>
> Danai Dima
>
|