Forecasting crude oil market volatility using extreme-value method
This study evaluates the forecasting performance of extreme-value volatility estimators in comparison with the GARCH volatility.In order to properly model the dynamics of volatility,two types of volatility models are discussed and estimated:return-based GARCH and extreme-value -based AV model.Examination of insample and out-of-sample volatility forecasts reveals that the AV model consistently outperforms the GARCH model.Our findings confirm that extreme-value volatility can retain its superiority in forecasting volatility by properly modeling the dynamic process.It would be beneficial to encompass intraday information especially price range to do volatility forecasting and risk management in crude oil future markets.
Volatility forecasting Extreme value GARCH Price range Crude oil future
Hongquan Li Shouyang Wang Fenghua Wen
Department of Finance,Changsha University of Science&Technology,Changsha,410076,China;Academy of Mat Academy of Mathematics and Systems Science,Chinese Academy of Sciences,Beijing,100190,China Department of Finance,Changsha University of Science&Technology,Changsha,410076,China
国际会议
长沙
英文
964-970
2008-10-28（万方平台首次上网日期，不代表论文的发表时间）