Analýza a predikce cen elektrické energie

Abstract

The subject of this master thesis is the market of electricity prices. The aim of this work is to analyze the current development of peak load spot prices in the Czech Republic from January 2012 to December 2021 and based on this development, find an adequate prediction model and estimate future prices development. First of all, the energy market and the factors affecting the level of prices are described. Then the theoretical basis for econometric modelling applying linear regression models are introduced. The main part of this thesis is focused to find an appropriate regression model for prediction. These models are verified and used for the prediction of further progression of peak load spot energy prices in the Czech Republic, with respect to an extreme price increase. It was found that the best is the autoregressive model with explanatory variables of natural gas prices in the Czech Republic and futures prices of European emission allowances. The data are overestimated compared to reality by only 2.98 % in case of using the model for ex-post prediction. The next part of thesis is devoted to the theoretical prediction of energy prices based on adjusted data without extreme price increases. Here again, the best is the autoregressive model with the prices of natural gas and emission allowances as explanatory variables. As the main reason for the difference between actual and predicted values can be considered the economic shocks caused by the coronavirus pandemic, by the reduction of the supply of commodities used for electricity production or by the Green Deal policy in the states of the European union.