Time-Series Forecasting Model for Evaluating Water Treatment Facility Efficiency in Uganda

Water treatment facilities in Uganda have been a focus of policy analysis to ensure safe drinking water for its population. However, there is limited empirical data on their efficiency over time. The analysis employs an autoregressive integrated moving average (ARIMA) model, which is robust for detecting trends and seasonal patterns in data. Uncertainty in forecasts is quantified using standard error estimates. In the period analysed, water treatment facilities showed a consistent improvement rate of approximately 5% per year, with no significant deviations from this trend. The ARIMA model effectively captured the efficiency trajectory, indicating that continued investment and monitoring are necessary to sustain these gains. Policy makers should prioritise ongoing maintenance and upgrades based on forecasted improvements to ensure long-term water quality. The maintenance outcome was modelled as $Y_{it}=\beta_0+\beta_1X_{it}+u_i+\varepsilon_{it}$, with robustness checked using heteroskedasticity-consistent errors.