Im Journal Power Electrical Systems – Extended Papers 2017 wurde mein Paper „Improving Short Term Load Forecasting with a Novel Hybrid Model Approach as a Precondition for Algorithmic Trading“ veröffentlicht. Das Paper umfasst Ansätze aus einigen Konferenzpapieren und gibt einen Ausblick zur Nutzung von optimierten Lastprognosen im automatisierten Handel an Intraday Märkten.
Hier der Abstract:
Considering the changing European power market, accurate electric load forecasts are of significant importance for power traders to reduce costs for ancillary services by leveling their position on continuous intraday power markets. The first part of the following case study, based on publicly available load data, focuses on a novel approach to combine different forecasting methodologies and techniques from the area of computational intelligence. The proposed hybrid model blends input forecasts from artificial neuronal networks, multi variable linear regression and support vector regression machine models with fuzzy sets to intraday and day-ahead forecasts. The forecasts are evaluated with commonly used metrics (mean average percentage error – MAPE & normalized rooted mean square error – NRMSE) to allow comparisons with other case studies.
Results from input forecasting models range from a yearly MAPE of 2.36% for the linear regression model to 2.1% for the support vector machine. Blended forecast from proposed hybrid models results in a MAPE of 1.46% for one hour and a MAPE of 1.72% for 24 hours ahead forecasts.
In the outlook section of the paper we show how to use blended forecasts as an input source for automatized intraday power trading with algorithms. Besides outlining use cases, model structure of trading algorithms and back testing approaches, the paper offers a state of the art insight on algorithmic trading in the power industry.