本文檢驗台灣股、匯市與美國股市之間是否存在傳導機制的危機蔓延效果、外溢效果,及個別市場的槓桿效果。首先利用Inclán and Tiao (1994)建議的疊代累積平方加總運算法(ICSS)檢測台灣股、匯市與美國股市報酬之變異多個結構性改變,在考量代表各國股票報酬變異之結構改變的虛擬變數下,估計一般化誤差分配GED-EGARCH模型來分析個別市場是否存在訊息不對稱的槓桿效果。另一方面,設定雙變量自我迴歸(VAR)模型,利用Granger因果關係檢定,檢測市場間報酬及變異是否存在外溢效果。最後,利用EGARCH模型所估計出的標準化殘差納入Engle(2002)建議之動態條件相關多變量GARCH模型,估計二市場報酬間隨時間變動的動態條件相關係數,以個別市場所發現的結構改變點劃分成多個區域,建立聯合信賴區間檢定市場間是否存在相關係數顯著改變的蔓延效果。實證結果顯示,台灣股、匯市及美國股市具有訊息不對稱的槓桿效果,而2國間衝擊蔓延的傳導機制則存在由美國股市外溢至台灣股市的蔓延效果。
In this paper, we are going to discuss the test example whether its consequence exists the effect of crisis contagion and spillover effect between the stock and exchange markets of Taiwan and the stock market of U.S., and also for the leverage effect of the individual market. The time points of structural changes in the volatility of the return are detected first, based on the iterated cumulative sums of squares (ICSS) algorithm developed by Inclán and Tiao (1994), and also adopts the generalized error distribution (GED)-EGARCH model to test the leverage effect, including the dummy variables which instead of the break points of the individual market. Second, this paper show that the VAR model can be used to test the spillover effect which use Granger-causality test. At last but not the least, this paper zeroes on the estimating the Dynamic conditional correlation-Multivariate GARCH models that was supported by Engle (2002), and also estimates the dynamic conditional correlation coefficient using the standard deviation that was estimated by GED-EAGRCH model. Moreover, this system also constructs the simultaneous confidence interval method to test contagion effect. Therefore, we find that the stock and exchange markets of Taiwan, and stock market of US exist the leverage effect, and the effect of crisis contagion from US spillover to Taiwan between two stock markets.
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。