SpatialEpiBench: Benchmarking Spatial Information and Epidemic Priors in Forecasting

2026-05-07

AI Summary

Existing epidemic forecasting methods struggle to effectively leverage spatial information under sparse, noisy, and non-stationary spatiotemporal data, and lack standardized evaluation benchmarks. To address this gap, this work proposes SpatialEpiBench—the first standardized spatiotemporal benchmark tailored for real-world epidemic prediction—encompassing 11 epidemiological datasets and incorporating rolling-window evaluation, geographic adjacency graph modeling, and embedding of epidemiological priors. Systematic evaluation reveals three critical failure modes: state-of-the-art models consistently underperform a simple “yesterday’s value” baseline across prediction horizons from one day to one month; even during outbreak periods, integrating epidemiological priors yields no significant performance gains; and current approaches exhibit fundamental limitations in early outbreak anticipation, robustness to noise, and spatial dependency modeling.

Abstract

Accurate epidemic forecasting is crucial for public health response, resource allocation, and outbreak intervention, but remains difficult with sparse, noisy, and highly non-stationary data. Because epidemics unfold across interacting regions, spatiotemporal methods are natural candidates for improving forecasts. Despite growing interest in spatial information, no standardized benchmark exists, and current evaluations often use simple chronological train-test splits that do not reflect real-time forecasting practice. We address this gap with SpatialEpiBench, a challenging benchmark for spatiotemporal epidemic forecasting in realistic public-health settings. SpatialEpiBench includes 11 epidemic datasets with standardized rolling evaluations and outbreak-specific metrics. We evaluate adjacency-informed forecasting models with widely used epidemic priors that adapt general models to epidemiology, but find that most methods underperform a simple last-value baseline from 1 day to 1 month ahead, even during outbreaks and with these priors. We identify three major failure modes: (1) poor outbreak anticipation, (2) difficulty handling sparsity and noise, and (3) limited utility of common geographic adjacency for epidemiological spatial information. We release benchmark data, code, and instructions at https://github.com/Rachel-Lyu/SpatialEpiBench to support development of operationally useful epidemic forecasting models.

Problem

• epidemic forecasting
• spatiotemporal modeling
• benchmarking
• spatial information
• epidemic priors

Innovation

• spatiotemporal forecasting
• epidemic modeling
• benchmarking
• spatial priors
• rolling evaluation