A convenient method used to discriminate between the two NB models is the comparison of the log-likelihood function values. Another standard method to distinguish the discussed models refers to information criteria, which is also based on the fitted loglikelihood function. Boucher et al. (2007) sustain that the criteria used to compare the models must penalize the one with a large number of parameters, considering the fact that the likelihood increases with the addition of parameters. The standard criteria refers to Akaike Information Criteria (