由於預測商品的波動越來越受到投資者的關切，本文擬利用 GARCH、GJR-GARCH、
QGARCH、EGARCH 作為波動模型，以台股期貨及美國 SPDR 自 2001年至2008年之日資料
作為實證標的，進行GARCH族模型的波動性預測能力評估。本研究以絕對報酬率、PK變幅、
GK 變幅、RS 變幅及已實現波動度(RV)作為市場真實波動性的代理變數，並同時採用對稱與
不對稱損失函數評估模型的波動性預測績效。首先，實證結果指出在絕大多數的情況下，不
對稱的波動模型較能解釋金融海嘯期間，金融市場的波動動態行為。其次，三種不對稱GARCH
族模型的預測績效互有領先，不過在大部分的情況，皆以 EGARCH模型最佳，GARCH 模型
表現最差。最後，以絕對報酬率或日變幅波動作為波動代理變數時，各模型的預測績效與已
實現波動的實證結果呈現相當一致的現象。因此，在進行台灣股價指數期貨及美國 SPDR 指
數型股票基金的波動性預測時，絕對報酬率及日變幅波動都是良好的波動替代變數。

The purpose of this study is to apply four GARCH-type models to daily volatility forecasting to the Taiwanese stock index futures and Standard & Poor’s Depository Receipts from 2001 to 2008. In stead of using squared returns as a proxy for true volatility, this study adopts absolute daily
returns, PK-range, GK-range, RS-range, and realized volatility, for use in the empirical exercise. The volatility forecast evaluation is conducted with a variety of volatility proxies according to both symmetric and asymmetric types of loss functions. Empirical results show that the EGARCH model provides the most accurate daily volatility forecasts, while the GARCH model performs the worst in general. Such evidence suggests that asymmetry in volatility dynamics should be taken
into account for forecasting financial markets volatility. Moreover, the latent volatility can be proxied using either absolute daily returns or daily price range with freely available prices.