Price Risk Measurement Based On Value-at-Risk Model In Electricity Market

The power industry is a basic industry of the national economy, and its operationalstatus has an important impact on the development of the national economy. From1982, the electricity market reform which start from Chile has a profound impact onthe world, and has formed some mature electricity market now. In the new electricitymarket environment, power trading as a commodity and its price determined bymarket competition. Relative to other markets, such as the traditional financialmarkets, the fluctuations of the electricity price return series is even more acutebecause the electrical energy can not storage. The days of the TOU can vary severaltimes, even though, the electricity prices can be negative, while different days,different months of the tariff is a difference of greater. The crisis of Californiaelectricity market make the two largest public utility power company loss$13billion,finally, the California government had spent billions of dollars of public funds anddistributed billions of dollars government bonds to help the power company to get ridof a serious crisis. Therefore, the volatility of electricity price make the electricitymarket face a huge risk, multi-level risk management contribute to the health andsustainable development of the electricity market,and has implications for our powermarket reforms.Given the dramatic electricity price volatility, we start from the inner electricitymarket, and control risk through the price measurement. Electricity price volatility hasa "fat tail" characteristics and the normal distribution can not describe the taildistribution, therefore, we introduce the extreme value theory to describe the yield tailof the distribution.In the mature PJM electricity market, we selected two moments,using the AR-GARCH model, the AR-EGARCH model and the AR-GARCH-POTmodel which mixed extreme value theory to measure the rate of return, and comparing the three models, in the end,we conclude that: Overall, theAR-GARCH-POT model measured the electricity market risk more accurate than theother two models. In particular, in a higher confidence level, such as the99%confidence level, the AR-GARCH-POT model has the least failing number, whichindicated that the AR-GARCH-POT model is more suitable to measure the fat tail ofthe data, and along with improved confidence, the AR-GARCH-POT model has ahigher accuracy. The accurate VaR has an important value to the risk control of theelectricity market.