Network Topology Matters, But Not Always: Mobility Networks in Epidemic Forecasting

TL;DR

This study evaluates the predictive value of mobility network structure versus volume for short-term COVID-19 forecasts in Massachusetts, finding network data most useful when local case histories are unavailable or delayed.

Contribution

It demonstrates when and how mobility network features improve epidemic forecasts, providing practical guidance on their use based on data availability.

Findings

Network features significantly improve forecasts without detailed case data.

Autoregressive models perform well when recent case data is available.

Network interactions are most valuable during epidemic waves and rapid changes.

Abstract

Short-horizon epidemic forecasts guide near-term staffing, testing, and messaging. Mobility data are now routinely used to improve such forecasts, yet work diverges on whether the volume of mobility or the structure of mobility networks carries the most predictive signal. We study Massachusetts towns (April 2020-April 2021), build a weekly directed mobility network from anonymized smartphone traces, derive dynamic topology measures, and evaluate their out-of-sample value for one-week-ahead COVID-19 forecasts. We compare models that use only macro-level incidence, models that add mobility network features and their interactions with macro incidence, and autoregressive (AR) models that include town-level recent cases. Two results emerge. First, when granular town-level case histories are unavailable, network information (especially interactions between macro incidence and a town's network position) yields large out-of-sample gains (Predict-R2 rising from 0.60 to 0.83-0.89). Second, when town-level case histories are available, AR models capture most short-horizon predictability; adding network features provides only minimal incremental lift (about +0.5 percentage points). Gains from network information are largest during epidemic waves and rising phases, when connectivity and incidence change rapidly. Agent-based simulations reproduce these patterns under controlled dynamics, and a simple analytical decomposition clarifies why network interactions explain a large share of cross-sectional variance when only macro-level counts are available, but much less once recent town-level case histories are included. Together, the results offer a practical decision rule: compute network metrics (and interactions) when local case histories are coarse or delayed; rely primarily on AR baselines when granular cases are timely, using network signals mainly for diagnostic targeting.