Electricity Price Prediction Using Multi-Kernel Gaussian Process Regression Combined with Kernel-Based Support Vector Regression

TL;DR

This paper introduces a hybrid model combining Gaussian Process Regression and Support Vector Regression to improve the accuracy of German electricity price predictions, effectively handling noise and outliers.

Contribution

The novel hybrid approach enhances GPR performance with SVR integration, providing more reliable out-of-sample predictions for electricity prices.

Findings

Outperforms benchmark models like LASSO autoregressive and deep neural networks.

Effectively manages noise and outliers in price data.

Improves prediction accuracy for German hourly electricity prices.

Abstract

This paper presents a new hybrid model for predicting German electricity prices. The algorithm is based on a combination of Gaussian Process Regression (GPR) and Support Vector Regression (SVR). Although GPR is a competent model for learning stochastic patterns within data and for interpolation, its performance for out-of-sample data is not very promising. By choosing a suitable data-dependent covariance function, we can enhance the performance of GPR for the German hourly power prices being tested. However, since the out-of-sample prediction is dependent on the training data, the prediction is vulnerable to noise and outliers. To overcome this issue, a separate prediction is calculated using SVR, which applies margin-based optimization. This method is advantageous when dealing with non-linear processes and outliers, since only certain necessary points (support vectors) in the training data are responsible for regression. The individual predictions are then linearly combined using uniform weights. When tested on historic German power prices, this approach outperforms the publicly available benchmarks, namely the LASSO estimated autoregressive regression model, deep neural network provided in the recent research by [1].