Bayes factor: A summary of evidence for the modelM1 against the another modelM0 provided by the set of data D, which can be used in the model selection. Given by the ratio of posterior to previous odds,
Twice the logarithm of B10 is on the similar scale as the deviance and the likelihood ratio test statistic. The following scale is many times useful for interpreting the values of B10;
Very sensitive to the assumed previous distribution of the models.