Sammendrag
The paper applies a Bayesian Estimator adapting MCMC simulation methodologies to build a general scientific stochastic volatility (SV) model for the mean and latent volatility of the base and peak load EEX one-year forward electric power contracts. The main objective is to find appropriate descriptions emphasizing schemes for portfolio management, risk management and general derivative pricing purposes. Moreover, as forecasting under the MCMC framework can be done easily for both the mean and volatility, the model building approach can produce useful and superior volatility and correlation updating schemes. The stochastic volatility model is based on a parameterized statistical model for simulation purposes and the estimation uses the MCMC simulation techniques. The methodology helps to circumvent the computational curse of dimensionality and is therefore superior to conventional derivative-based hill climbing optimizers. Our results show that the general scientific methodology from the Bayesian model parameter estimation, adequately describes the European energy market's financial contracts. The successful implementation to energy markets suggests not whether the methods can be used in financial market applications, but how efficient the methods can generally become.
Vis fullstendig beskrivelse