 |
 |
Dear DCM experts I have a few questions on how to read the output of DCM post-hoc. I hope you can help. 1. If I understand correctly, DCM.Pp is the model posterior for family partitioning, comparing the family with vs. without each parameter. If this is true: a. Why is DCM.Pp only 1's and 0's rather than any other value between 0-1 ? Is it thresholded? b. What can the user specified families (DCM.fun and DCM.Pf) be used for? How is it different from DCM.Pp? 2. Given that DCM.Ep contains parameter estimates under the selected model – why does it have values for parameters that have been removed in the selected model? 3. What determines the number of reduced models shown in the graphic output of model posteriors? Why is the wining model always the last one? I'd appreciate your help! Tali Bitan University of Haifa