The Institut Louis Bachelier has just published the eighth issue of its “ILB Methods” series, which presents the quantitative models used in the academic world. This issue is devoted to probabilistic forecasts on energy markets. It has been produced with the participation of Olivier Féron, researcher at EDF R&D and scientific director of the Energy Markets Finance Laboratory (FiME).

Since the opening up of the European electricity market in the early 2000s, price forecasting has become an essential activity for all players in the sector. Spot forecasting is the most widely used method. It uses statistical analysis and artificial intelligence to forecast price trends on the spot market (24 hours).

Margin of error and price volatility

However, performance remains limited, with a margin of error of 10 to 15%. In addition, the development of renewable energies and recent crises have increased the volatility of energy prices.

As a result, probabilistic forecasting, which takes uncertainty into account, has recently gained in popularity. Unlike point forecasts, probabilistic forecasts estimate a range of possible outcomes with the associated probabilities.

Watch the video below for a detailed explanation of probabilistic forecasting models and the advantages of this approach.

 

Summary note