From: Quantile regression for overdispersed count data: a hierarchical method
Quantile | Model 1 | Model 2 | ||||
---|---|---|---|---|---|---|
0.05 | 0.5 | 0.95 | 0.05 | 0.5 | 0.95 | |
Fit | ||||||
 Log predictive density (lpd) | −27,742 | −26,643 | −27,391 | −27,733 | −26,644 | −27,379 |
 Complexity (pwaic) | 3387 | 3764 | 3723 | 3385 | 3771 | 3725 |
 WAIC | 62,258 | 60,815 | 62,229 | 62,235 | 60,830 | 62,209 |
Model Checks | ||||||
 Predictive Coverage (% of observations with 95% CRI of yrep including observation) | 0.966 | 1.000 | 0.975 | 0.968 | 1.000 | 0.976 |
Posterior Predictive p tests | ||||||
 Log likelihood ratio | 0.17 | 0.45 | 0.24 | 0.17 | 0.43 | 0.25 |
 Maximum observation | 0.47 | 0.44 | 0.47 | 0.46 | 0.43 | 0.50 |
 Sum of observations | 0.50 | 0.51 | 0.51 | 0.49 | 0.49 | 0.53 |