| Summary: | 碩士 === 銘傳大學 === 財務金融學系碩士班 === 98 === The studies have mostly measured by the standard deviation of close prices in
Previous. We use time series（like ARIMA、GARCH）to catch leptokurtic and fat tail of
Reward of Financial assets. Parkinson (1980) forcefully argues and demonstrates the
superiority of using range as a volatility estimator as compared with standard methods.
Then , Chou（2002）provided the Conditional Autoregressive Range Model (CARR)
which is the range Combined with the GARCH mondel.In this experimental , the
CARR model outperforms than GARCH model both in in-sample and out-of sample
forecasts of weekly stock market volatilities.
In recent years, the option of stock index in Taiwan , its trading volume was growing
rapidly, and the option price is closely related to its volatility, and Taiwan Futures
Exchange has been officially available VIX index for investors. So, the purpose of this
study is to establish the Optimal Forecasting Modeling in the VIX Index of TXO. This
applies the following six single models, such as ARIMA model, ARIMAX model,
ARIMA-GARCH model, ARIMAX-GARCH model, ARIMA-CARR model,
ARIMAX-CARR model. Compare real volatility index with the above six models and
them forecast performance.
By comparing the forecasting performance of the volatility form the above six
models mentioned, ARIMAX-CARR model which adds Exogenous（the rate of change of
volume of TWSE） ranked the best.
|
|---|
