On Retention and Growth Patterns of Abuja Electricity Distribution Company Metered Customers through Times Series Forecasting.

Abstract

This study investigates the long term growth dynamics of metered customers within the Abuja Electricity Distribution Company (AEDC) from 2015 to the first quarter of 2024, using descriptive analytics, trend modelling, and time series forecasting. The descriptive analysis shows a sustained upward trajectory in metering coverage, characterized by policy driven surges during key roll out periods. A linear trend model reveals a highly significant positive slope β = 4, 946.69 and strong explanatory power (R2 = 0.94), confirming that time is a dominant driver of metering expansion. Annual retention rates further highlight substantial growth in several years, with values exceeding 110% during intensive metering campaigns. In order to forecast future metering performance, multiple ARIMA models were compared, and ARIMA (1, 1, 1) emerged as the optimal specification based on lowest AIC, BIC, RMSE, and MAPE values. The model’s Ljung-Box p-value of 1 indicates that residual autocorrelation is fully addressed which validate its suitability for forecasting. Projections for 2024–2028 indicate continued growth under current operational and policy conditions. Overall, the findings demonstrate strong historical progress and provide a statistically sound basis for future planning and metering investment strategies within AEDC.