Prediction of electricity prices

The aim of this paper is to analyse time series that might have an impact on prices of energy commodities. Analysis and econometric modelling reveal which time series are corelated and how. This is input for further modelling of forecasting tool for electricity and gas prices (in near future). Selected time series are electricity and gas prices and contracted quantities per day of futures for next year on several European markets, oil prices, coal prices, prices of emission allowances, weather (temperature and sunlight defining the consumption of energies), prices of electricity on spot market, currency exchange rate (EUR/CZK and EUR/USD) and correlation of particular week day. For the predicted value was chosen price of electricity and gas on PXE - Czech commodity stock exchange (part of EEX – German stock that operates with energy commodity markets of almost all EU member states). The product is future “Cal+1” (product delivery is in next year) and time series were selected for period 1/2016-10/2018. These days experts predict prices based of several indicators based on technical and fundamental (their expert opinion with minimum number outcomes), but there is no exact model with nor exact prediction tool. This model is nowadays valuable more than ever due to rising prices and volatility (in year 2018 prices of electricity in Czech Republic steeply increased from 34 EUR/MWh to 58 EUR/MWh (peak in 9/2018) and the market was more volatile than ever (7 EUR/MWh decrease in 4 days). Also, gas prices reaching 27 EUR/MWh are at their historical maximum. So far there is only few public researches covering this topic (and only for electricity spot prices) and only few econometric methods are applied. That is why is this research so unique, not mentioning the benefit for public sector, where the model might generate enormous profits.