• Media type: E-Book; Doctoral Thesis; Electronic Thesis; Text
  • Title: Application of Structural Electricity Models - From Parameter Estimation and Parameter Risk to an Implied Hedging Framework
  • Contributor: Harms, Cord [Author]
  • imprint: University of Duisburg-Essen: DuEPublico2 (Duisburg Essen Publications online), 2017-06-26
  • Language: English
  • Keywords: hedging ; parameter risk ; structural model ; electricity ; parameter estimation ; Fakultät für Mathematik
  • Origination:
  • Footnote: Diese Datenquelle enthält auch Bestandsnachweise, die nicht zu einem Volltext führen.
  • Description: This thesis provides a mathematical analysis of structural electricity spot models and their applicability for dynamic hedging in practice. The analysis requires the application of stochastic analysis, computational finance and asymptotic statistics. Due to the complexity to store electricity, it usually has to be consumed immediately after its production. Consequently, seasonalities in electricity wholesale prices are not only observed between different months (due to weather conditions) or between different weekdays (due to industry and retail demand) but also between different hours of a day (i.e. due to retail demand and the photovoltaic power production). In detail, they are caused by fundamentals, i.e. expected demand, expected marginal\footnote{The most expensive power plant in the grid which is necessary to serve the demand sets the overall wholesale price.} power plants and expected weather conditions (i.e. intensity of the sun, or expected wind conditions). As soon as fundamentals behave unexpectedly, the electricity (equilibrium) price has to change in order that demand and production volumes can rematch. The close relation between electricity prices and fundamental information has been the basic idea to investigate so-called structural electricity models. These models integrate fundamental knowledge of the price setting mechanism into an electricity spot model. However, they still ensure close-form formulas for at least forward contracts. Though structural models are well investigated in terms of derivatives pricing, there is still a lack of applicability in practice, i.e. how those models can be used for hedging and how model parameters are estimated respectively. The basic idea of our research is that a structural electricity spot model implies an electricity forward model (between fuels and electricity) being close to a model with cointegrated forwards. The dynamic is directly implied by the merit order of the market. Furthermore, electricity forwards become risk-neutral $\Q$-martingales under ...
  • Access State: Open Access