Bayesian Neural Networks with Monte Carlo Dropout for Probabilistic Electricity Price Forecasting
It addresses uncertainty in electricity price forecasting for energy market decision-makers, but is incremental as it applies an existing method (MC dropout) to a specific domain.
This work tackled the problem of electricity price forecasting by developing a Bayesian neural network with Monte Carlo dropout for probabilistic predictions, which outperformed benchmark models like GARCHX and LEAR in terms of point prediction and intervals.
Accurate electricity price forecasting is critical for strategic decision-making in deregulated electricity markets, where volatility stems from complex supply-demand dynamics and external factors. Traditional point forecasts often fail to capture inherent uncertainties, limiting their utility for risk management. This work presents a framework for probabilistic electricity price forecasting using Bayesian neural networks (BNNs) with Monte Carlo (MC) dropout, training separate models for each hour of the day to capture diurnal patterns. A critical assessment and comparison with the benchmark model, namely: generalized autoregressive conditional heteroskedasticity with exogenous variable (GARCHX) model and the LASSO estimated auto-regressive model (LEAR), highlights that the proposed model outperforms the benchmark models in terms of point prediction and intervals. This work serves as a reference for leveraging probabilistic neural models in energy market predictions.