What are you trying to do? Test a hypothesis? Or perhaps you've already tested a hypothesis using Bayesian Model Selection, and now you want report the parameters of a winning model?
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of Toshi Kawagoe
Sent: 15 February 2017 00:45
To: [log in to unmask]
Subject: Re: [SPM] different DCM results between simple averaging and BPA averaging
Thank you for your reply.
I tried it and found that it produced closer result with arithmetic average, although it slightly different.
Then, should not I use the arithmetic average and use the BPA ?
Is it completely wrong to use arithmetic average?