Interdisciplinary perspectives on infectious diseases · 2025
Forecasting the Monkeypox Outbreak Using Limited Data: A Case Study of Thailand.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Shuaib Sherif Eneye, Muangprathub Jirapond, Intarasit Arthit, Riyapan Pakwan
Аннотация
INTRODUCTION: On August 14, 2024, the World Health Organization (WHO) declared monkeypox (mpox) a Public Health Emergency of International Concern (PHEIC). Shortly thereafter, Thailand reported Asia's first confirmed case of the Clade Ib strain. Given Thailand's high international travel volume, anticipating short-term case trajectories is essential for situational awareness.
METHODS: We forecast monthly mpox cases using the Department of Disease Control (DDC) data from July 2022 to September 2024 and evaluate Poisson and negative binomial generalized linear models (GLMs), Holt-Winters exponential smoothing, NeuralProphet, and a stacked ensemble. Rolling-origin cross-validation was used to assess out-of-sample accuracy (MAE, RMSE, and MAPE) and empirical interval coverage (80% and 95%).
RESULTS: Holt-Winters provided the most accurate single-model forecasts by RMSE, while the stacked ensemble yielded the most reliable uncertainty calibration. Forecasts for October-December 2024 suggest relatively low case counts with non-negligible uncertainty, indicating the need for continued vigilance.
DISCUSSION: These findings are exploratory decision-support inputs that complement ongoing surveillance and may help inform adaptive planning under both low- and high-incidence scenarios.
METHODS: We forecast monthly mpox cases using the Department of Disease Control (DDC) data from July 2022 to September 2024 and evaluate Poisson and negative binomial generalized linear models (GLMs), Holt-Winters exponential smoothing, NeuralProphet, and a stacked ensemble. Rolling-origin cross-validation was used to assess out-of-sample accuracy (MAE, RMSE, and MAPE) and empirical interval coverage (80% and 95%).
RESULTS: Holt-Winters provided the most accurate single-model forecasts by RMSE, while the stacked ensemble yielded the most reliable uncertainty calibration. Forecasts for October-December 2024 suggest relatively low case counts with non-negligible uncertainty, indicating the need for continued vigilance.
DISCUSSION: These findings are exploratory decision-support inputs that complement ongoing surveillance and may help inform adaptive planning under both low- and high-incidence scenarios.