Dear SPM/Karl
Thanks for the updated version of spm_expm.m This certainly helped in the
particular convergence problem we were having, but as you may have
suspected even though this loop was no longer a problem the DCM as a whole
converged poorly. The PEB iterations in particular began to exceed 32
iterations and eventually began returning warnings that matrices were
singlular (spm_PEB lines 223 which is iCE = inv(Ce); and 229 which is
Cby=inv(XX'*iCeX) ).
What does this mean about the DCM? Does this say something about the
regions or connections being too related or too unrelated?
Darren
-------------------------------------------------------------------------
Darren R. Gitelman, M.D.
Cognitive Neurology and Alzheimer¹s Disease Center
Northwestern Univ., 320 E. Superior St., Searle 11-470, Chicago, IL 60611
Voice: (312) 908-9023 Fax: (312) 908-8789
-------------------------------------------------------------------------
|