雙變量GARCH-DCC模型應用在美國對亞洲四小龍股票市場之研究

美國股市居全球領導地位，為國際傳導現象的主要影響來源，美國股市的變動可以迅速傳遞至國外，進而引起其他股市的連動。本研究以亞洲四小龍股市為實證對象，進行其受美國股市的股價及市場報酬之傳遞波動影響分析，使用雙變量GARCH-DCC模型，採自2000年1月1日至2011年12月31日共2685筆日資料，進行分析。實證得到以下結論：美國與亞洲四小龍之股價沒有通過共整合檢定，顯示沒有長期均衡關係。Granger因果關係檢定顯示美國市場報酬影響台灣、新加坡及南韓三個國家市場報酬，且呈現單向因果關係；美國與香港則具雙向關係。美國與亞洲四小龍股票市場的日報酬資料皆有ARCH效果。GARCH模型顯示亞洲四小龍均受美國前一期市場報酬正向影響，本國當期股價的波動會受到前期未預期變動(新訊息)與前期報酬波動(舊訊息)的正向影響，但受到美股前期未預期變動的負向影響。GARCH-DCC之估計與分析進一步顯示亞洲四小龍之股票市場報酬受美國前一期報酬正向影響，當期報酬波動亦受美股前期未預料報酬波動正向影響，但受美股前期預期報酬波動的負向影響。

關鍵字

雙變量GARCH-DCC模型；共整合檢定； Granger因果關係檢定

並列摘要

The stock market of United State (US) keeps in leading level for many years. Its volatility has extremely influence on the stock market of other foreign countries. A bivariate dynamic conditional correlation GARCH model (DCC model) was applied to investigate the relationship of stock index and stock returns between US and four Asian stock markets separately, which includes Taiwan, South Korea, Singapore and Hong Kong. In this study, totally 2685 daily data of stock index and stock return were collected and analyzed from January 1, 2000 to December 31, 2011. Empirical results showed that no long-term correlation of stock index and stock returns between US and four Asian countries. The result of Granger causality indicated that the unilateral positive contemporaneous relationships exist between US stock return and intraday stock returns of Taiwan, Singapore and South Korea. However, bilateral relationships existed in the volatility of stock returns between US and Hong Kong. The GARCH model showed that the intraday stock return of US has positive volatility driven effect on four Asian countries. Furthermore, the GARCH-DCC analysis proved that the stock returns of four Asian countries were influenced by intraday stock returns of US.