 |
 |
Dear Drew, This is the posterior covariance of the parameter estimates - how uncertain the algorithm is about the parameters. Its used by other DCM functions to, for example, compute the probability that the connection is larger than some threshold (eg. 0). If this is DCM for fMRI use the 'review' option here. Its computed using the Variational Laplace algorithm see eg. equation 7 in [1] To look at posterior variances of each parameter broken down into fields - you can do Cp=spm_unvec(diag(DCM.Cp),DCM.Ep); full(Cp.A) full(Cp.B) etc. See also spm_dcm_estimate for list of fields. Best, Will. [1] K.Friston, J. Mattout, N. Trujillo-Barreto, J. Ashburner, and W. Penny. Variational free energy and the Laplace approximation. Neuroimage, 34(1):220-234, 2007. From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of Sevel,Landrew S Sent: 11 February 2013 19:54 To: [log in to unmask] Subject: [SPM] DCM.Cp DCM experts, I have some difficulty understanding the output for this value. What is the utility of this field and how are its values determined? Best and many thanks, Drew