{ "background": "Time-series forecasting models are critical for the strategic planning and financial management of water treatment infrastructure. A previously published model for assessing the cost-effectiveness of such systems in a major West African nation required independent validation to assess its robustness and generalisability. ", "purpose and objectives": "This study aimed to replicate and critically evaluate the methodological rigour and predictive performance of the specified forecasting model. The objective was to verify its computational reproducibility and test its validity using an updated, extended dataset. ", "methodology": "A direct computational replication was performed, followed by a measurement replication using an expanded dataset from treatment facilities. The core model, an autoregressive integrated moving average (ARIMA) framework expressed as \ (B) (1-B) ᵈ yt = \ (B) \, was re-implemented. Predictive accuracy was assessed via mean absolute percentage error (MAPE) and Theil's U statistic, with uncertainty quantified using 95% confidence intervals for all forecasted values. ", "findings": "The replication confirmed the model's structural form but revealed a systematic overestimation of cost-effectiveness by approximately 12% in the medium-term forecasts. While the model remained stable (Theil's U < 1), the confidence intervals for key operational cost parameters were notably wider than originally reported, indicating greater predictive uncertainty. ", "conclusion": "The original model is methodologically sound but requires recalibration to correct for optimistic bias in its predictions. The wider confidence intervals suggest the model's parameters may be sensitive to specific temporal or regional data characteristics. ", "recommendations": "Future applications should incorporate recalibrated coefficients and explicit uncertainty bounds into decision-making. Model utility would be enhanced by integrating non-stationary covariates reflecting fluctuating energy and chemical costs. ", "key words": "replication study, time-series analysis, cost forecasting, water treatment, infrastructure management, ARIMA modelling", "contribution statement": "This study provides the first independent validation and critical methodological evaluation of a prominent forecasting model for water treatment economics,
C. A. C. Okonkwo (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: