Leave-one-out cross validation model comparison using the estimated difference in ELPD (expected log pointwise predictive probabilities).

ELPD is a measure of out-of-sample predictive accuracy, as estimated by the Bayesian leave-one-out cross validation (LOO). We compared model P (cases as primary) to models PS and S (primary and secondary cases and secondary cases only). Models under the horizontal bar at 0 fit the data better than model P (reference). Vertical bars represent the corresponding standard error of the difference in ELPD.