Pricing and risk management for highly cyclical commodity markets
- Publication Type:
- Issue Date:
This thesis is concerned with electricity as a commodity, as traded in markets such as the Australian National Electricity Market (NEM). While there are many different forms of electricity market in the world, the NEM exhibits a very good example of several levels of cyclicity resulting from the fundamentally physical nature of the market. The variations of price from the ‘normal’ daily cycle are as rapid and extreme as the reversions to this cycle. The three main reasons why electricity in the NEM exhibits such extreme behaviour are: • The NEM is fundamentally a physical, rather than financial market; • Electricity cannot be stored; and • Supply must be maintained. In this thesis we take three approaches to model the demand and price and price derivatives based on electricity markets. Each approach is explained in a dedicated chapter. They are: 1. We model demand using a series of lognormal distributions, fitted to the quantiles of a market forecast; 2. We forecast price using a structural model which is based on the Least Action Principle used in mechanics. 3. We price options using two models, both based on a series of distributions around each time period. (a) For caps/floors on price, we model distributions around the expected price curve and back out implied volatilities for each time period within the contract; and (b) For options on load (or output), we model distributions around the expected demand curve and use our implied volatilities from the earlier demand curve to calibrate these. Each of the three chapters builds on the previous one - the price curve described in chapter 4 requires an analytical representation of demand which is derived in chapter 3. The option pricing described in chapter 5 requires the price curve (for price caps) and the demand curve (for volumetric caps) that is derived in the previous chapter. Together these separate models build to form the beginnings of an integrated pricing and risk management suite.
Please use this identifier to cite or link to this item: