Cross-Country Learning for National Infectious Disease Forecasting Using European Data

TL;DR

The paper tackles the challenge of limited national historical data for infectious disease forecasting by introducing a cross-country learning framework that trains a single model on pooled time series from multiple countries and evaluates it on a target country, demonstrated through Cyprus COVID-19 case forecasting with European data. It systematically analyzes the impact of lookback window length and cross-country data augmentation, comparing Transformer, XGBoost, and baseline methods. Key findings show that incorporating data from other countries yields consistent improvements over national-data models and baselines, with a 14-day lookback providing robust performance when using all available cross-country data. The approach offers a practical pathway to improve short-term outbreak forecasts in data-constrained settings and can be extended to other infectious diseases and surveillance targets such as hospitalizations and ICU admissions.

Abstract

Accurate forecasting of infectious disease incidence is critical for public health planning and timely intervention. While most data-driven forecasting approaches rely primarily on historical data from a single country, such data are often limited in length and variability, restricting the performance of machine learning (ML) models. In this work, we investigate a cross-country learning approach for infectious disease forecasting, in which a single model is trained on time series data from multiple countries and evaluated on a country of interest. This setting enables the model to exploit shared epidemic dynamics across countries and to benefit from an enlarged training set. We examine this approach through a case study on COVID-19 case forecasting in Cyprus, using surveillance data from European countries. We evaluate multiple ML models and analyse the impact of the lookback window length and cross-country `data augmentation' on multi-step forecasting performance. Our results show that incorporating data from other countries can lead to consistent improvements over models trained solely on national data. Although the empirical focus is on Cyprus and COVID-19, the proposed framework and findings are applicable to infectious disease forecasting more broadly, particularly in settings with limited national historical data.

Figures (4)

• Figure 1: Cyprus' COVID-19 cases per day (Period: 2020-06-01 to 2022-12-31 for visual purposes).
• Figure 2: 7-day moving averages of confirmed cases across countries.
• Figure 3: Training and test sets splits used in the experiments.
• Figure 4: Spearman's correlation matrix.