Dear KV
There's no right answer here, but as heuristic, look for at least 10% explained variance.
Best
Peter
-----Original Message-----
From: SPM (Statistical Parametric Mapping) <[log in to unmask]> On Behalf Of SUBSCRIBE SPM KVoigt
Sent: 29 October 2019 07:02
To: [log in to unmask]
Subject: [SPM] DCM model convergence statistics
Hi,
I’m using spm_dcm_check_fmri() in order to check whether my first level dcm models converged.
The diagnostics report: variance explained, largest parameter estimate, number of estimated parameters, posterior correlations of the parameters.
I would like to know whether there are any guidelines on for what patterns / values I should be looking for in order to be able to judge whether my models have converged.
Thank you very much in advance,
KV
|