Seemingly unrelated time series model for forecasting the peak and short-term electricity demand: Evidence from the Kalman filtered Monte Carlo method

Прогнозування Пікового Навантаження На Електричні Мережі

“…Time series forecasting, including the energy sector, uses a variety of methods. These methods can be classified into three categories: statistical methods, machine learning methods, and deep learning methods [4]. Among the statistical methods considered are ARIMA, SARIMA and ETS, which are based on the analysis of trends and seasonality.…”

“…This approach offers enhanced prediction accuracy, scalability for large grids, and adaptability to varying conditions. Moreover, the article [4] delves into the application of Kalman and filtered Monte Carlo methods for load forecasting. By analyzing unlinked time series models, the authors forecast peak and total electricity demand.…”

“…Furthermore, an ensemble of models was taken into account, encompassing Random Forest, a technique that amalgamates multiple models to enhance prediction accuracy [4]. Moreover, specific libraries tailored for time series analysis were employed, such as Neural Prophet, leveraging neural networks for predictive purposes.…”