Modelling monkeypox outbreak using exponential smoothing models

Monkeypox has re-emerged as a global health concern, especially in the Democratic Republic of Congo (DRC), where monthly case trends show a worrying rise (WHO, 2022). Accurate forecasting of outbreak patterns is essential for public health planning and timely interventions. This study aims to model and forecast monkeypox cases using exponential smoothing methods. Monthly case data from January 2023 to February 2025 was collected from the World Health Organization and preprocessed to correct outliers using the Interquartile Range (IQR) method and median imputation. Five models Naïve, Mean, Simple Exponential Smoothing, Holt’s Linear Trend, and Brown’s Double Exponential Smoothing were evaluated using Repeated Time Series Cross-Validation (Aziz et al., 2018). The Holt model achieved the best performance, with a MAPE of 18.28% and a six-month forecast accuracy of 81.7%, proving it effective for short-term outbreak prediction. These findings highlight the practical application of exponential smoothing in disease forecasting and support its use in improving public health readiness and response strategies