Temporal Trends, Inequality and Forecasting of ISP Latency in Nigeria, 2020–2023

Internet Service Providers (ISPs) experience variability and inequality in network latency, affecting service quality and user experience. This study addresses assessment of latency performance by quantifying trends, inequality, regime changes, and forecasting behaviour across ISPs. Monthly latency measurements collected from January 2020 to December 2023 were cleaned by removing zero and negative observations and summarised using descriptive statistics. Gini and Theil indices and Lorenz curves quantified latency inequality, while seasonal-trend decomposition and Augmented Dickey-Fuller test examined behaviour and stationarity. A two-state Markov-switching autoregressive model classified network conditions into “good” and “bad” regimes. ARIMA, Holt-Winters, Prophet and Random Forest models were evaluated using CRPS and pinball loss metrics. Mean latency ranged from 8.04 ms for Dimension Data to 32.73 ms for InqDigital. Random Forest achieved the best performance. The study demonstrates that combining inequality analysis, regime classification and machine-learning forecasting can support early identification of elevated latency and stable ISP QoS.