Out-of-sample crash prediction on City Crash and Exposure Data (San Francisco held-out)
-6.63Predictive Log-LikelihoodIndep RE
Evaluation results
| Method | Links | |
|---|---|---|
| Indep RERandom-effects prior=independent hierarchical2026.06 | -6.63 | |
| GP EuclideanRandom-effects prior=Gaussian Process, Kernel type=Euclidean covariate distance2026.06 | -6.76 | |
| GP learnedRandom-effects prior=Gaussian Process, Kernel type=learned embedding2026.06 | -6.95 | |
| GP HDV-onlyRandom-effects prior=Gaussian Process, Kernel type=log-HDV-frequency similarity2026.06 | -7.04 | |
| Single poolRandom-effects prior=none2026.06 | -16.99 |