Dear Fabienne,
Calculate a BMS separately in each group -ensure you switch on Bayesian Model Averaging (BMA) over your models for each group (it's an option in the batch). Then take the estimates produced by the BMA (the fields of the BMA are documented somewhere in the SPM manual) and compare them between groups using your favourite statistics software.
Let us know if you need more specific guidance.
Best,
Peter
-----Original Message-----
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of Fabienne Jung
Sent: 10 July 2015 15:12
To: [log in to unmask]
Subject: [SPM] Inference on parameters - DCM
Dear all,
I have several questions regarding the analysis of connectivity parameters in DCM since I am not sure how to proceed with my data:
I have calculated DCMs for all my subjects and did RFX BMS. Now, I'd like to investigate the strength of the connections of my network.
1.) I'd like to estimate how strong left and right primary motor cortex are connected and compare these values between two groups.
2.) Furthermore, I'd like to compare two connections with two other connections (a lateral to a medial pathway). Is it allowed to just sum up values of two forward connections?
I don't know if I explained everything in an understandable way... My problem is that I can not decide on whether to do this analysis on all models or just on one winning model.
Thanks in advance for your help!
Best
Fabienne
|