Electricity Price Forecasting: Bridging Linear Models, Neural Networks and Online Learning

• URL: http://arxiv.org/abs/2601.02856v1
• Date: Tue, 06 Jan 2026 09:35:02 GMT
• Title: Electricity Price Forecasting: Bridging Linear Models, Neural Networks and Online Learning
• Authors: Btissame El Mahtout, Florian Ziel,
• Abstract summary: We propose a novel neural network approach that combines linear and nonlinear feed-forward neural structures.<n>Compared to the current benchmark models, the proposed forecasting method significantly reduces computational cost.<n>Our results are derived from a six-year forecasting study conducted on major European electricity markets.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Precise day-ahead forecasts for electricity prices are crucial to ensure efficient portfolio management, support strategic decision-making for power plant operations, enable efficient battery storage optimization, and facilitate demand response planning. However, developing an accurate prediction model is highly challenging in an uncertain and volatile market environment. For instance, although linear models generally exhibit competitive performance in predicting electricity prices with minimal computational requirements, they fail to capture relevant nonlinear relationships. Nonlinear models, on the other hand, can improve forecasting accuracy with a surge in computational costs. We propose a novel multivariate neural network approach that combines linear and nonlinear feed-forward neural structures. Unlike previous hybrid models, our approach integrates online learning and forecast combination for efficient training and accuracy improvement. It also incorporates all relevant characteristics, particularly the fundamental relationships arising from wind and solar generation, electricity demand patterns, related energy fuel and carbon markets, in addition to autoregressive dynamics and calendar effects. Compared to the current state-of-the-art benchmark models, the proposed forecasting method significantly reduces computational cost while delivering superior forecasting accuracy (12-13% RMSE and 15-18% MAE reductions). Our results are derived from a six-year forecasting study conducted on major European electricity markets.

Related papers

• Recurrent Neural Networks with Linear Structures for Electricity Price Forecasting [0.0]
  We present a novel recurrent neural network architecture designed explicitly for day-ahead electricity price forecasting.<n>Our combined forecasting model embeds linear structures, such as expert models and Kalman filters, into recurrent networks.<n>The proposed model achieves approximately 12% higher accuracy than leading benchmarks.
  arXiv  Detail & Related papers  (2025-12-04T11:38:30Z)
• A Comparative Study of Machine Learning Algorithms for Electricity Price Forecasting with LIME-Based Interpretability [8.114643944442296]
  This study compares eight machine learning models using Spanish electricity market data.<n>Results show that KNN achieves the best performance with R2 of 0.865, MAE of 3.556, and RMSE of 5.240.
  arXiv  Detail & Related papers  (2025-12-01T02:51:41Z)
• Bayesian Neural Networks with Monte Carlo Dropout for Probabilistic Electricity Price Forecasting [0.0]
  This work presents a framework for probabilistic electricity price forecasting using Bayesian neural networks (BNNs) with Monte Carlo (MC) dropouts.<n>A critical assessment and comparison with the benchmark model, namely: generalized autoregressive conditional heteroskedasticity with a variable (GARCHX) and the LASSO estimated auto-regressive model (LEAR)
  arXiv  Detail & Related papers  (2025-11-12T13:33:28Z)
• Data-driven jet fuel demand forecasting: A case study of Copenhagen Airport [43.17090130312271]
  We evaluate the performance of data-driven approaches using a substantial amount of data obtained from a major aviation fuel distributor in the Danish market.<n>Our analysis compares the predictive capabilities of traditional time series models, Prophet, LSTM sequence-to-sequence neural networks, and hybrid models.<n>To ensure the reliability of the data-driven approaches and provide valuable insights to practitioners, we analyze three different datasets.
  arXiv  Detail & Related papers  (2025-11-04T11:45:58Z)
• Adaptive Online Learning with LSTM Networks for Energy Price Prediction [0.5514902789425196]
  This study focuses on developing a predictive model to forecast day-ahead electricity prices in the California energy market.<n>The model incorporates a variety of features, including historical price data, weather conditions, and the energy generation mix.<n>The results demonstrate that the custom loss function can improve the model's performance, aligning predicted prices more closely with actual values.
  arXiv  Detail & Related papers  (2025-10-19T15:48:38Z)
• From Distributional to Quantile Neural Basis Models: the case of Electricity Price Forecasting [42.062078728472734]
  We introduce the Quantile Neural Basis Model, which incorporates the interpretability principles of Quantile Generalized Additive Models.<n>We validate our approach on day-ahead electricity price forecasting, achieving predictive performance comparable to distributional and quantile regression neural networks.
  arXiv  Detail & Related papers  (2025-09-17T15:55:59Z)
• Forecast-PEFT: Parameter-Efficient Fine-Tuning for Pre-trained Motion Forecasting Models [68.23649978697027]
  Forecast-PEFT is a fine-tuning strategy that freezes the majority of the model's parameters, focusing adjustments on newly introduced prompts and adapters. Our experiments show that Forecast-PEFT outperforms traditional full fine-tuning methods in motion prediction tasks. Forecast-FT further improves prediction performance, evidencing up to a 9.6% enhancement over conventional baseline methods.
  arXiv  Detail & Related papers  (2024-07-28T19:18:59Z)
• Generative Probabilistic Time Series Forecasting and Applications in Grid Operations [47.19756484695248]
  Generative probabilistic forecasting produces future time series samples according to the conditional probability distribution given past time series observations. We propose a weak innovation autoencoder architecture and a learning algorithm to extract independent and identically distributed innovation sequences. We show that the weak innovation sequence is Bayesian sufficient, which makes the proposed weak innovation autoencoder a canonical architecture for generative probabilistic forecasting.
  arXiv  Detail & Related papers  (2024-02-21T15:23:21Z)
• End-to-End Reinforcement Learning of Koopman Models for Economic Nonlinear Model Predictive Control [45.84205238554709]
  We present a method for reinforcement learning of Koopman surrogate models for optimal performance as part of (e)NMPC. We show that the end-to-end trained models outperform those trained using system identification in (e)NMPC.
  arXiv  Detail & Related papers  (2023-08-03T10:21:53Z)
• A Hybrid Model for Forecasting Short-Term Electricity Demand [59.372588316558826]
  Currently the UK Electric market is guided by load (demand) forecasts published every thirty minutes by the regulator. We present HYENA: a hybrid predictive model that combines feature engineering (selection of the candidate predictor features), mobile-window predictors and LSTM encoder-decoders.
  arXiv  Detail & Related papers  (2022-05-20T22:13:25Z)
• Predictive Accuracy of a Hybrid Generalized Long Memory Model for Short Term Electricity Price Forecasting [0.0]
  This study investigates the predictive performance of a new hybrid model based on the Generalized long memory autoregressive model (k-factor GARMA) The performance of the proposed model is evaluated using data from Nord Pool Electricity markets.
  arXiv  Detail & Related papers  (2022-04-18T12:21:25Z)
• The impact of online machine-learning methods on long-term investment decisions and generator utilization in electricity markets [69.68068088508505]
  We investigate the impact of eleven offline and five online learning algorithms to predict the electricity demand profile over the next 24h. We show we can reduce the mean absolute error by 30% using an online algorithm when compared to the best offline algorithm. We also show that large errors in prediction accuracy have a disproportionate error on investments made over a 17-year time frame.
  arXiv  Detail & Related papers  (2021-03-07T11:28:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.