On modelling the electricity futures curve

Haar, Rick ter (2010) On modelling the electricity futures curve.

Abstract:The electricity industry has changed considerably during the last decade. With new market participants entering the market of former monopolists and the introduction of derivatives, it has become increasingly important to develop accurate price models for these contracts, both for risk management and valuation purposes. In this study we statistically analyze the German electricity future market. Research particularly aimed at modelling the EEX futures curve is still scarce, while innovative new approaches have been developed. This research creates an overview of the different approaches and selects a model that is the best candidate to model the German electricity futures curve. With the illiquidity of the EEX option market in mind, the following research question was formulated: Which price models are best suited to model the German electricity futures curve, taking into account our wish to have closed-form option pricing formulas? Based on our study of the (German) electricity market and the performed data analysis on the EEX futures prices, we require good futures curve models to: • Include seasonal patterns. Prices are higher for contracts delivering electricity during winter months. • Allow the specification of a complex volatility structure. Volatility depends on the length of the delivery period, the time to delivery and the time of the delivery period. Accurately modelling this complex structure is critical, also for option pricing. • Create a good fit to the initially observed futures curve. • Produce analytical, closed-form option price formulas. Based on these requirements, we consider the simplified direct swap model proposed by Benth & Koekebakker (2008) to be the best candidate. The main advantages of this model are that: • The model does not rely on a non-explicit relation between spot and futures prices nor on smoothing algorithms to derive the futures price dynamics. Only observations of actually traded futures contracts are used. • A perfect fit with initial futures curve is created. • It is possible to specify the complex volatility structure and still have analytical, closedform option pricing formulas. Drawbacks of the approach are: • Only market data of the futures contracts that cannot be decomposed into smaller contracts can be used (atomic swaps). • We cannot infer the spot price dynamics from the swap price dynamics. • We need a lognormal specification of the atomic swaps in order to have analytical, closed-form option pricing formulas. It is shown that this specification cannot fully capture the fat tails of the log-returns.
Item Type:Essay (Bachelor)
Nuon N.V.
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:85 business administration, organizational science
Programme:Industrial Engineering and Management BSc (56994)
Link to this item:http://purl.utwente.nl/essays/60867
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page