Out-of-Sample RMSPE Values for Spatiotemporal Models of California Mortality

5.5 KB of out-of-sample Root Mean Square Percentage Error (RMSPE) values for a novel spatiotemporal model and reference models. The data accompanies a paper modeling monthly respiratory-related deaths across California, using social deprivation indices and air quality surrogate variables. The dataset was authored by Jeffrey Wu and last updated on 2026-05-21.

Use Cases

• Comparing forecast accuracy between a novel KGR-SKATER model and reference spatiotemporal models based on RMSPE values.
• Evaluating model performance in non-stationary time series settings based on the synthetic case study results mentioned.
• Benchmarking spatiotemporal forecasting methods for high-dimensional, non-Gaussian data as described in the paper.

Strengths

• Data is openly available under a CC-BY-4.0 license.
• The associated paper provides a detailed methodological context for the model evaluation.

Limitations

• The dataset is very small at 5.5 KB, indicating limited scope.
• Row count and specific column definitions are unknown, requiring inspection after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Jeffrey Wu, via figshare and Dryad repository.

Collection Method

Generated from model evaluation as part of a research paper proposing a spatiotemporal modeling procedure.

Freshness

Last updated 2026-05-21 07:31:27; freshness should be verified.

Geography

Likely California, based on the described case study of respiratory deaths in the state.