本文以Chan and Maheu (2002) 所提出ARJI (Autoregressive Conditional Jump Intensity) 模型，將外生變數放入平均數方程式(mean equation)及條件變異數(conditional variance)方程式中，即為ARJI-X 模型，進一步探討上、下變幅以及成交量、未平倉量等相關因子對於亞洲各股價指數期貨-日經225股價指數期貨(NKX)、台灣證交所加權股價指數期貨(TWX)、摩根台指期貨(MSTWX)、南韓綜合指數期貨(KMX)、吉隆坡綜合股價指數期貨(IKX)與香港恆生股價指數期貨(HSX)之報酬率及條件變異數的影響。實證結果發現，上下變幅確實對報酬率及條件變異數有著顯著不同的影響；此外，在成交量及未平倉量的變動率因子上，落後一期的成交量變動率對條件變異數之影響為負，而未平倉量變動率亦為負向之影響，顯示未平倉量的變動可反映出市場深度的改變。因此，本文將市場上的價格與成交量訊息加以整理而得的上下變幅及成交量變動因子經由ARJI-X模型，更能有效的詮釋其與報酬率及其波動性的關係，而使以更深入的觀點來看待市場中的各項資訊。 This study applies ARJI-X models which entering up, down range and other related factors into the return and conditional variance equation of ARJI model, proposed by Chan and Maheu (2002), to capture the dynamics of volatility on Asian stock index futures markets by allowing volatility to depend on both volume effects and other related information. The empirical result shows that both up and down range have significant and different effects on return and conditional variance. It is also found of a negative effect of lag one period’s volume rate of change and open interest rate of change on volatility. Altogether, the ARJI-X model is more appropriate than traditional statistical models because it is capable of interpreting observed statistical characteristics of many time series of financial assets.
