Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/30184
Year of Publication: 
2010
Series/Report no.: 
IWQW Discussion Papers No. 02/2010
Publisher: 
Friedrich-Alexander-Universität Erlangen-Nürnberg, Institut für Wirtschaftspolitik und Quantitative Wirtschaftsforschung (IWQW), Nürnberg
Abstract: 
In this article we present both a theoretical framework and a solved example for pricing an European gas storage facility and computing the optimal strategy for its operation. As a representative price index we choose the Dutch TTF day-ahead gas price. We present statistical evidence that the volatility of this index is time-varying, so we introduce a new continuous-time model by incorporating GARCH diffusion into an Ornstein-Uhlenbeck process. Based on this price process we use dynamic programming methods to derive partial differential equations for pricing a storage facility. As an example we apply our methodology to a storage site located in Epe at the German-Dutch border. In this context we investigate the effects of multiple contract types, and perform a sensitivity analysis for all model parameters. We obtain a value surface displaying the properties of a financial straddle. Both volatility and mean reversion influence the facility value - but only around the long-run mean of the gas price. The terminal condition, which includes information about the contract provisions, is of importance if it contains e.g. penalty terms for low inventory levels. Otherwise its influence is diminishing for increasing lease periods.
Subjects: 
TTF gas price
GARCH diffusion
natural gas storage
dynamic computing
JEL: 
C31
C61
Document Type: 
Working Paper

Files in This Item:
File
Size
343.58 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.