Deep Generative Spatiotemporal Engression for Probabilistic Forecasting of Epidemics

TL;DR

This paper introduces deep spatiotemporal engression models for probabilistic epidemic forecasting, effectively capturing complex dependencies and providing reliable uncertainty quantification, outperforming existing benchmarks across multiple datasets.

Contribution

The paper presents a novel deep generative framework for probabilistic epidemic forecasting that incorporates endogenous uncertainty and establishes theoretical properties like ergodicity and stationarity.

Findings

Outperforms several benchmarks in point and probabilistic forecasts

Demonstrates consistent accuracy across six epidemiological datasets

Provides explainability to support public health decision-making

Abstract

Accurate and reliable forecasting of epidemic incidences is critical for public health preparedness, yet it remains a challenging task due to complex nonlinear temporal dependencies and heterogeneous spatial interactions. Often, point forecasts generated by spatiotemporal models are unreliable in assigning uncertainty to future epidemic events. Probabilistic forecasting of epidemics is therefore crucial for providing the best or worst-case scenarios rather than a simple, often inaccurate, point estimate. We present deep spatiotemporal engression methods to generate accurate and reliable probabilistic forecasts on low-frequency epidemic datasets. The proposed methods act as distributional lenses, and out-of-sample probabilistic forecasts are generated by sampling from the trained models. Our frameworks encapsulate lightweight deep generative architectures, wherein uncertainty is quantified endogenously, driven by a pre-additive noise component during model construction. We establish geometric ergodicity and asymptotic stationarity of the spatiotemporal engression processes under mild assumptions on the network weights and pre-additive noise process. Comprehensive evaluations across six epidemiological datasets over three forecast horizons demonstrate that the proposal consistently outperforms several temporal and spatiotemporal benchmarks in both point and probabilistic forecasting. Additionally, we explore the explainability of the proposal to enhance the models' practical application for informed, timely public health interventions.