以各種模型及避險策略研究股價指數期貨之避險效益

The Hedging Effectiveness of Stock Index Futures for Alternative Models and Hedging Strategies

本研究以台灣加權、香港恆生HS、日本Nikkei225、美國道瓊工業、S&P500、NASDAQ、英國FTSE100、法國CAC40、及德國DAX30等指數期貨契約為研究樣本，分別在極大化預期效用與極小化投資組合風險的目標函數下，對不同風險趨避程度之投資者，以各種動態避險模型及傳統靜態避險模型，探討避險期間不對稱雙變量誤差修正GJR-EC-GARCH模型、雙變量GJR-GARCH模型、EC-OLS模型、OLS模型、及單純避險的避險績效之差異，以供各國投資人進行股價指數期貨避險時之重要依據。經由實證後發現在避險績效方面，各模型在不同風險趨避程度下大多能有效地提高投資者之預期效用或降低其投資組合風險。而在各模型間避險效益分析上，不論風險趨避程度為何，均以誤差修正模型的避險績效為最佳。在模型的選擇上，對於積極型的投資人而言，採用GJR-EC-GARCH模型可獲得較佳的避險績效，至於保守型的投資人而言，則可採GJR-EC-GARCH或EC-OLS模型進行避險。

This paper investigates the hedging effectiveness of TAIFEX Taiwan, Hang Seng, Nikkei225, DJIA, S&P500, NASDAQ, FTSE100, CAC40, and DAX30 stock index futures for investors with different degrees of risk aversion, under the objective functions of maximizing the expected utility and minimizing the portfolio risk. The hedging performance is measured and compared by using the various dynamic and static hedging models, including the asymmetric bivariate GJR-EC-GARCH, GJR-GARCH, EC-OLS, OLS, and Naïve models. The empirical results show that most of the hedging models can increase the expected utility or decrease portfolio risk effectively. Moreover, the error-correction models outperform the other models in terms of hedging performance for different types of investors. The hedging performance obtained by the GJR-EC-GARCH model is better for the less risk-averse investors, while the hedging effectiveness of the GJR-EC-GARCH and EC-OLS models is higher for the investors with higher degrees of risk aversion.