Time-Series Forecasting Model Evaluation of Public Health Surveillance Systems in Uganda,

Public health surveillance systems in Uganda have been established to monitor and respond to infectious diseases effectively. A time-series forecasting model was developed using historical data from Uganda's public health surveillance system. The model accounts for seasonal variations in disease incidence through an autoregressive integrated moving average (ARIMA) approach. Uncertainty in the model predictions is quantified using robust standard errors and confidence intervals. The ARIMA model demonstrated a strong predictive ability, forecasting influenza-like illness trends with a mean absolute error of 5% and a 95% confidence interval indicating that the model's forecasts are likely within ±20% of actual values. The time-series analysis revealed seasonal patterns in disease incidence. The ARIMA model provided reliable predictions for public health responses, highlighting the importance of continuous system evaluation to ensure effective risk reduction strategies. Regular model updates and validation against new data are recommended to maintain the accuracy and relevance of the forecasting models within Uganda's surveillance systems. Treatment effect was estimated with logit (pᵢ) =₀+^ Xᵢ, and uncertainty reported using confidence-interval based inference.