 |
 |
Dear Klass and SPMers,
In claculating group Bays Factor, for many subjects, I do have this message in matlab command window. this is especially for those 2 models I am most interested to compare.
AIC and BIC disagree about which model is superior - no decision can be made.
out of 11 subjects I do only have 4 subject who have a value for this 2 model comparison.
so I was wondering what to do now? looks like overall, I can not have a strong evidence in favor of one model.
However looking at only BIC, there is strong evidence in favor of one for almost 8 subejcts.
Would it be valid to argue only based on BIC approximation?
Many thanks,
Mina
|