| Summary: | 碩士 === 淡江大學 === 財務金融學系碩士班 === 94 === Volatility forecasting is very important to derivative pricing, hedging, and risk management. This paper using GARCH, GJR-GARCH models and the VIX index of TAIEX Options to compare their forecasting ability.
The empirical evidence show that using daily data to forecast the performance, GJR-GARCH model is superior, while using intraday data, the explanatory power of all models are obviously enhanced and errors are also improved. Therefore, we approve that the more intensive data can obtain the higher explanatory power, lower forecasting errors and more information content.
In summary, we find that the VIX index of TAIEX Options using intraday data seems an unbiased estimator to forecast the real volatility, although it does not have the best performance than other models in MAE and RMSE testing indicators.
|
|---|
