本文依Brooks (2001)。所提之單變數雙門檻GARCH模型及Glosten,Jaganathan與Runkle (1993)所提之GJR-GARCH模型之想法,提出一個雙變數之雙門檻-IGRACH模型,利用此模型探討成交量與匯率波動率對南韓綜合股價指數報酬之波動的影響,且以成交量與匯率波動率的正負值作為門檻,其研究資料期間是採用1999年1月4日到2006年12月28日的南韓綜合股價指數、成交量與匯率的資料。而由實證結果顯示AR(5)-雙門檻-IGRACH(1,2)模型對探討成交量與匯率波動率對南韓股票市場報酬的影響是合適的,且反應出南韓股票市場具有不對稱的效果。而由實證結果也顯示在成交量與匯率波動率正負值的四種組合之下,成交量與匯率波動率的信息將會影響股票市場報酬的波動,且反應出成交量與匯率波動率也將會影響股票市場報酬波動的變異風險。
This paper combines the idea of simple variable double GARCH model (Brooks, 2001) and the idea of GJR-GARCH model (Glosten, Jaganathan and Runkle, 1993) to propose a double threshold-IGRACH model of the two variables. We u se this model to discuss the influence of the turnover volume and the exchange rate volatility rate on the South Korea stock price return volatility, and take the threshold by the turnover volume and the exchange rate volatility rates' positive and negative value. The empirical diagnosis results demonstrate that the AR(5)-double threshold-GRACH(1,2) model is appropriate to discusses the turnover volume and the exchange rate volatility rate to the South Korea stock market return influence, and it also responds that the South Korea stock market has an asymmetrical effect. The empirical diagnosis result also shows that the information about the turnover volume and the exchange rate volatility rate actually affects the stock market return volatility by the combination of the positive and negative value for the turnover volume and the exchange rate volatility rate. This result also shows that the turnover volume and the exchange rate volatility rate can affect the variation risk of the stock market return volatility.