Zusammenfassung:
Digital infrastructure is a crucial factor for inclusive growth, and rural digital service quality, especially in developing economies, is still behind that of cities. Therefore, continuous monitoring and short-term forecasting of service quality is important for effective management of digital infrastructure. This study presents a data-based approach for monitoring rural internet service quality in Uzbekistan, with average speed measured in Mbps being considered a fundamental service quality parameter. The study focuses on rural areas in the Samarkand region, with internet speeds measured monthly over a sample period from March 2022 to December 2025, totalling 46 observations. The objectives of this study include: (i) establishing the existence of a significant trend in rural internet speeds, (ii) establishing whether there is an annual seasonal effect, and (iii) establishing whether short-run forecasts are suitable for operations. The study employed a deterministic linear trend equation, a unit root test with seasonal differencing, and a SARIMA model, with Stata being employed for estimation purposes. The results indicate that there is a strong, significant trend in rural internet speeds, with an average increase of about 0.48 Mbps per month, with β = 0.484, p < 0.01. The study also indicates that there is no complete removal of non-stationarity by employing a seasonal difference, with a p- value of 0.4775, indicating that there is a long-term development trend. Using the results obtained, a proposed automated monitoring logic is presented, with trend and forecast deviations being considered signals for service quality. This study provides a useful approach towards developing monitoring solutions that will be useful for monitoring rural IT service quality, especially in developing economies.
