Efficient Leave-One-Out Cross-Validation and WAIC for Bayesian
Models
Description
We efficiently approximate leave-one-out cross-validation (LOO)
using very good importance sampling (VGIS), a new procedure for regularizing
importance weights. As a byproduct of our calculations, we also obtain
approximate standard errors for estimated predictive errors, and for the
comparison of predictive errors between two models. We also compute the
widely applicable information criterion (WAIC).