Forecasting Moroccan Electricity Imports Using Time Series Models: A Comparative Study of SARIMA and Support Vector Regression

This study investigates Morocco’s monthly electricity import data to evaluate forecasting models.The objective is to assess the forecasting accuracy of two time series models—SARIMA (Seasonal AutoRegressive Integrated Moving Average) and SVR (Support Vector Regression) time series models using data from 2008 to 2021. The goal is to identify the most suitable model for predicting electricity imports and to understand its potential. Although SVR is widely considered a leading forecasting model in the literature, this study examines its computational aspects, including parameter selection, alongside a detailed comparison of fforecast accuracy with SARIMA. The results of this research may also be applied to forecast other economic time series, such as electric power exports and the trade of petroleum products, including gasoline, oil, and gas.