Modeling the Prevalence of Malaria Using Autoregressive Integrated Moving Average: Evidence from the University of Ilorin Teaching Hospital, Nigeria

Malaria continues to impose a substantial health and economic burden in sub-Saharan Africa, particularly in Nigeria, despite decades of prevention and control interventions. This study employs a time series modeling approach to analyze monthly malaria incidence data from the University of Ilorin Teaching Hospital (UITH) covering January 2013 to December 2023. The Augmented Dickey–Fuller test confirmed stationarity in the dataset, while model selection based on Akaike Information Criterion (AICc) identified ARIMA (3,0,3) as the optimal model. Diagnostic validation using residual plots and the Box–Ljung test confirmed model adequacy. Forecasts for 2024–2025 indicate an average monthly incidence of approximately 44 cases, with minor fluctuations in early 2024 stabilizing in subsequent months. The findings highlight the utility of statistical forecasting for anticipatory public health planning, enabling evidence-based resource allocation, targeted interventions, and enhanced readiness for seasonal surges. Integrating predictive analytics into national malaria surveillance frameworks could strengthen efforts toward achieving sustained disease reduction and eventual elimination.