Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/53777
標題: The Cointegration Analysis and the Economic Drivers of Dynamic Conditional Correlation
動態條件相關之共整合與經濟趨動因素分析
作者: 董澍琦
關鍵字: 應用研究;Dynamic Conditional Correlation;經濟學;動態條件相關;共整合;EGARCH 模型;Cointegration;EGARCH-DCC model.
摘要: 
現代投資理論指出,風險降低的程度取.於各股票市場間的相關性,然而許多實證研究發現,各資產報酬的相關係數會隨時間而變,因此,如要建構一最佳化之投資組合,除了過去重視的平均相關係數之外,更要深刻瞭解其動態條件相關。最近許多學者發展出精細的統計模型,用以描述相關係數隨時間而變的特性。雖然各統計模型都有其優點,但卻無法透露任何訊息,到底是什.因素造成相關係數的改變。因此,本多年期研究計畫的重點就在於尋找總體經濟因素以解釋動態條件相關。第一年度計畫將採用EGARCH-DCC 模型探討亞洲股票市場與美國股票市場之間的訊息傳遞效果,從模型中可得到動態條件相關的時間序列數據,接著我們進行共整合分析以檢測亞洲各國與美國動態條件相關是否具有共整合關係。第二年度計晝,將建立一實證模型,根據財務金融理論及過去實證結果,尋找影響動態條件相關的可能經濟因素或特殊事件。第三年度計畫仍是使用相同實證模型(EGARCH-DCC),探討股票市場與外匯市場之間的訊息傳遞效果,接著得到股票報酬與匯價變動的動態條件相關數據,然後尋找影響相關係數變動的因素。

The degree of risk reduction will highly depend on the correlations betweeninternational stock markets. However, the empirical studies have shown that thecorrelations between equity returns vary considerably through time. To construct anoptimal portfolio at each point time, it is not only important to have an insight into theaverage level of the correlations, but in particular into their dynamics, i.e. the conditionalcorrelations. Therefore, several sophisticated statistical models are developed to explain thetime-variation of conditional correlations. Although these statistical models have theirattractive features, they do not reveal any information about the economic sources drivingthe conditional correlations. The 「search for economic sources」will be the main purpose ofthis project. The first-year job of this project is to apply the EGARCH-DCC model toinvestigate the return and volatility transmission behavior between US stock return and anyAsian stock return. Then we get the time-series data of the dynamic conditional correlationcoefficient through the DCC model. By using the cointegration analysis, we try toinvestigate whether there are common factors among these dynamic conditionalcorrelations. The second-year job attempts to explain why correlations vary through timeby relating this time-variation to some economic variables. In other words, in thesecond-year, we will search for the fundamental sources driving the conditionalcorrelations. The third-year job employs the same methodology to investigate the dynamicconditional correlations between the stock market and the foreign exchange market. Wealso want to search for fundamentals that may influence the degree of comovementbetween stock returns and the change of exchange rates. The 『search for economic forces orfind the common trends」makes this project radically different from most other studies ontime-varying correlations, which were mainly concerned with modeling correlations in apurely statistical way.
URI: http://hdl.handle.net/11455/53777
其他識別: NSC96-2416-H005-015-MY2
Appears in Collections:財務金融學系所

Show full item record
 

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.