Data on future price and volume development, like future prices for power, gas, coal and oil, with a maximum of reliability is an essential factor of success for the optimization of complex generation and procurement portfolios in the energy industry.
Every commodity price has its own specific characteristics with seasonal, weekly, and daily patterns, its periodically changing volatilities, and its mean reversion properties which must be precisely modeled. Commodity prices often considerably correlate with each other on short-term, long-term and medium-term time scales, both within one market area and on different markets. All these properties can be mapped into a mathematical (stochastic) price model.
Parameter Estimation
Stochastic processes often contain different parameters, such as volatility, mean reversion or correlations to other processes
To adapt a stochastic process on the real developments, it is crucial to estimate the parameters of the stochastic process as well as possible
We use mathematical estimation algorithms based on the maximum likelihood method
Scenario Generation
Using the estimated parameters of the selected stochastic processes Monte Carlo scenarios are generated
They can be used to generate Monte Carlo scenarios, which are, for example, used for risk assessment of open spot positions or schedules
These price scenarios are generated risk neutral with respect to the price forward curve, e.g. to the HPFC of the EEX spot price
Integration
Components of DT.Analytics for the estimation of stochastic parameters and processes can be integrated into the optimization models of the DT.Energy Suite
Alternatively, DT.Analytics is available as a stand-alone modeling and analysis tool
More detailed information about our modeling can be found in the brochure.
