University of Technology, Sydney

Value at Risk for Australian Electricity Markets

Email address: tony.hall@uts.edu.au

Keywords: Electricity markets; Risk measures; Value at risk; Extreme value theory

JEL Classifications: C.22; G.10; G.21

Abstract:
Since the deregulation of the electricity markets and the establishment of the Australian National Electricity Market in December 1998, spot electricity prices have been characterised by periods of extreme prices, which quickly revert to normal levels. This price volatility has led to concerns over the capital adequacy of generators and other market participants. A common tool for measuring risk exposures is Value at Risk (VaR). The methodology used in this study is to apply extreme value theory to estimate the tail of the innovation distribution of spot prices. VaR computed from this approach is contrasted with the traditional measures for computing VaR. As different market participants have time horizons that range from a half hour through to a calendar month, risk measures over a range of time intervals are required. With the availability of only two years of market data (measured half hourly), an issue of particular interest is the possibilty of scaling the VaR measures from the frequency of observation up to more highly aggregated time periods. A modified “bootstrap” method suggested by McNeil and Frey (2000) is adapted to obtain VaR measures over multiple horizons, and these measures are contrasted with those obtained using square-root scaling laws.

PDF file of paper: Not available.

Session: Asset Prices II

Time: Sunday, 8 July, 8am - 9:30am

Room: C