We propose a novel machine learning approach for probabilistic forecasting of hourly day-ahead electricity prices. In contrast with the recent advances in data-rich probabilistic forecasting, which approximates distributions with few features (such as moments), our method is nonparametric and selects the distribution from all possible empirical distributions learned from the input data without the need for limiting assumptions. The model that we propose is a multioutput neural network that accounts for the temporal dynamics of the probabilities and controls for monotonicity using a penalty. Such a distributional neural network can precisely learn complex patterns from many relevant variables that affect electricity prices. We illustrate the capacity of the developed method on German hourly day-ahead electricity prices and predict their probability distribution via many variables, doing so more accurately than the state-of-the-art benchmarks can, thus revealing new valuable information in the data.
Economic information from Smart Meter: Nexus Between Demand Profile and Electricity Retail Price Between Demand Profile and Electricity Retail Price
Stanford University(斯坦福大学);GEIRI North America(通用电气研究院北美分部);Beijing Electric power Economic Technology Research Institute(北京电力经济技术研究院);Toulouse School of Economics(图卢兹经济学院)
Comparing the Forecasting Performances of Linear Models for Electricity Prices with High RES Penetration
Free University of Bozen-Bolzano(博尔扎诺自由大学);London Business School(伦敦商学院);BI Norwegian Business School(挪威商业管理学院);Vrije Universiteit Amsterdam(阿姆斯特丹自由大学)
Department of Information Technology and Electrical Engineering, ETH Zürich(苏黎世联邦理工学院信息与电气工程学院);Department of Economics and Nuffield College, University of Oxford(牛津大学经济学系与纳菲尔德学院)