本研究經由Bai and Perron (2003)的結構轉變設定並採用新的ARJI-Trend模型來探討美國道瓊工業指數與史坦普爾500指數之條件異數的恆常成份與移轉成份以及跳躍的強度與頻率之間的關係。結果發現條件變異數的恆常成份與移轉成份的確存在，並且在發生重大事件期間不但條件變異數之恆常成份會有增加的現象，而移轉成份增加的幅度更大。跳躍強度亦在條件變異數之移轉成份發生較大變化時大幅增加，並且可以看出條件變異數之移轉成份不但影響跳躍發生的機率，也同樣影響跳躍的強度。

This paper studies the relationship between the permanent and transitory components of the conditional variance and the frequency and intensity of jump of return. We use a new ARJI-Trend model to capture the daily of American Dow Jones Industrial Index and S&P500 Index with a structural break analysis by Bai and Perron (2003). We find that both permanent and transitory comonents of the conditional variance are really exist in the whole sample period. The permanent component increase during the event occurrence and the transitory component increase larger at the same time. Jump intensity also makes a large change as the transitory component of the conditional variance increase when abnormal event occur.