Journal
INTERNATIONAL JOURNAL OF FORECASTING
Volume 39, Issue 2, Pages 884-900Publisher
ELSEVIER
DOI: 10.1016/j.ijforecast.2022.03.001
Keywords
Deep learning; NBEATS and NBEATSx models; Interpretable neural network; Time series decomposition; Fourier series; Electricity price forecasting
Categories
Ask authors/readers for more resources
We introduce NBEATSx, an enhanced version of the NBEATS model that incorporates exogenous variables to improve forecast accuracy in electricity price prediction. The NBEATSx model outperforms the original NBEATS model and other established statistical and machine learning methods by achieving a nearly 20% increase in forecast accuracy. Additionally, the NBEATSx model provides interpretability through its ability to decompose time series and visualize the impacts of trend, seasonality, and exogenous factors.
We extend 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 tasks across a broad range of years and markets. We observe state-of-the-art performance, significantly improving the forecast accuracy by nearly 20% over the original NBEATS model, and by up to 5% over other well-established 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. To assist related work, we made the code available in a dedicated repository.(c) 2022 The Author(s). Published by Elsevier B.V. on behalf of International Institute of Forecasters. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available