Enhanced time series forecasting using hybrid ARIMA and machine learning models

Abstract

Accurate energy demand forecasting is essential for optimizing resource management and planning within the energy sector. Traditional time series models, such as ARIMA and SARIMA, have long been employed for this purpose. However, these methods often face limitations in handling nonstationary data, complexity in model tuning, and susceptibility to overfitting. To address these challenges, this study proposes a hybrid approach that integrates traditional statistical models with advanced computational methods. By combining the strengths of both approaches, the proposed models aim to enhance predictive accuracy, improve computational efficiency, and maintain robustness across varied energy datasets. Experimental results demonstrate that these hybrid models consistently outperform standalone traditional methods, providing more reliable and precise forecasts. These findings underscore the potential of hybrid methodologies in advancing energy demand forecasting and supporting more effective decision-making in energy management.