Closed M2vaine closed 8 years ago
I have solved the mystery. The problem came from my former version of the toolbox (which never was on github). In VBA_logDet instead of the condition indIn = find(~isinf(dq)&dq~=0); the absurd condition: indIn = find(~isinf(dq)&dq<=eps); was applied.
So the good news is that all git-hub versions of toolbox are free from this error.
I have been rerunning some inversion with the toolbox. The posteriors, fits data are the same, only the F values (initial and final) are different. They differ quite a lot on an individual subject (~4), and at the level of the group it completely changes the results. What have been changed in the computaiton of F?