Investigation of forecasting methods for the hourly spot price of the day-ahead electric power markets

Abstract

Forecasting hourly spot prices for real-time electricity usage is a challenging task. This paper investigates a series of forecasting methods to 90 and 180 days of load data collection acquired from the Iberian Electricity Market (MIBEL). This dataset was used to train and test multiple forecast models. The Mean Absolute Percentage Error (MAPE) for the proposed Hybrid combination of Auto Regressive Integrated Moving Average (ARIMA) and Generalized Linear Model (GLM) was compared against ARIMA, GLM, Random forest (RF) and Support Vector Machines (SVM) methods. The results indicate significant improvement in MAPE and correlation co-efficient values for the proposed hybrid ARIMA-GLM method.