Research on risk measure of electricity market based on armax-garch model with conditional skewed-t distribution and extreme value theory

Abstract

How to effectively evaluate price of volatility risk is the basis of risk management in electricity market. An ARMAX-GARCH model imposing a skewedt-t distribution with time-varying skewness and degree of freedom over the error terms (ARMAX-GARCH-ST) is proposed and used to filter electricity price series in order to capture the dependencies, seasonalities, heteroscedasticities, skewnesses, leptokurtosises, volatility-clustering and relationship to system loads. In this way, an approximately independently and identically distributed residual series with better statistical properties is acquired. Then Extreme Value Theory (EVT) is adopted to explicitly model the tails of the normalized residuals of ARMAX-GARCH-ST model and accurate estimates of electricity market Value-at-Risk (VaR) can be produced. The empirical analysis shows that the ARMAX-GARCH-EVT models can be rapidly reflect the most recent and relevant changes of spot electricity prices and can produce accurate forecasts of VaR at all confidence levels, showing better dynamic characteristics. These results present several potential implications for electricity markets risk quantifications and hedging strategies. © 2013 Asian Network for Scientific Information.