Dear Susanne,
Yes that's fine. If you had a family which was the clear winner, you might just want to generate an average model (BMA) from the winning family. Otherwise, choose all families.
Best,
Peter
-----Original Message-----
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of Susanne Dietrich
Sent: 26 March 2015 12:18
To: [log in to unmask]
Subject: [SPM] DCM family comparisons
Dear SPM/DCM users,
I would like to use BMS model family comparisons. Is it common practice to define families which do not contain the same number of models?
If yes, how to deal with such families?
Further, specifying family for BMA, is it common practice to average across "all families" or the "winning family"?
Kind regards
Susanne
--
Dr. Susanne Dietrich
Dept. of Neurology / MEG-Center
University of Tübingen
Otfried-Müller-Str. 47
72076 Tübingen
phone: ++49 (0)7071-29 87708 priv.: 0163 695 4814
fax: ++49 (0)7071-29 5706
email: [log in to unmask]
|