Table 4. Validation statistics for the partial and covariate Bayesian models.

Model	AUC*	Mean error†	Mean absolute error‡
Partial (1a)	0.71	0.21	6.23
Partial (1b)	0.68	0.42	6.86
Covariate model	0.78	0.18	4.27

* AUC values in the range of 0.5–0.7 indicates poor discriminative capacity, 0.7–0.9 is considered good and > 0.9 to be very good.

^† Overall tendency to over or under-predict relative risk.

^‡ Overall precision of models estimated using magnitude of error in predictions.