Neural basis expansion analysis with exogenous variables: Forecasting electricity prices with NBEATSx

04/12/2021

∙

We extend the neural basis expansion analysis (NBEATS) to incorporate exogenous factors. The resulting method, called NBEATSx, improves on a well performing deep learning model, extending its capabilities by including exogenous variables and allowing it to integrate multiple sources of useful information. To showcase the utility of the NBEATSx model, we conduct a comprehensive study of its application to electricity price forecasting (EPF) tasks across a broad range of years and markets. We observe state-of-the-art performance, significantly improving the forecast accuracy by nearly 20 the original NBEATS model, and by up to 5 statistical and machine learning methods specialized for these tasks. Additionally, the proposed neural network has an interpretable configuration that can structurally decompose time series, visualizing the relative impact of trend and seasonal components and revealing the modeled processes' interactions with exogenous factors.

READ FULL TEXT

Neural basis expansion analysis with exogenous variables: Forecasting electricity prices with NBEATSx

Neural networks in day-ahead electricity price forecasting: Single vs. multiple outputs

Short- and long-term forecasting of electricity prices using embedding of calendar information in neural networks

N-BEATS: Neural basis expansion analysis for interpretable time series forecasting

Forecasting day-ahead electricity prices: A review of state-of-the-art algorithms, best practices and an open-access benchmark

Predictive Accuracy of a Hybrid Generalized Long Memory Model for Short Term Electricity Price Forecasting

Forecasting electricity prices with machine learning: Predictor sensitivity

Neural basis expansion analysis with exogenous variables: Forecasting electricity prices with NBEATSx

Related Research

Neural networks in day-ahead electricity price forecasting: Single vs. multiple outputs

Short- and long-term forecasting of electricity prices using embedding of calendar information in neural networks

N-BEATS: Neural basis expansion analysis for interpretable time series forecasting

Forecasting day-ahead electricity prices: A review of state-of-the-art algorithms, best practices and an open-access benchmark

Predictive Accuracy of a Hybrid Generalized Long Memory Model for Short Term Electricity Price Forecasting

Forecasting electricity prices with machine learning: Predictor sensitivity