透過您的圖書館登入
IP:3.231.55.243
  • 學位論文

跳躍風險與波動傳遞效果-原油、不動產、黃金與匯率之實證研究

Jump Risk and Volatility Transmission Effects between Crude Oil, REITs, Gold and Exchange

指導教授 : 邱建良
共同指導教授 : 張鼎煥(Ting-Huan Chang)

摘要


本研究探討自2005年以來原油價格、不動產、黃金現貨價格及美元匯率四者之互動關係。樣本期間為2005年6月17日至2011年9月30日之德州輕原油收盤價、美國不動產基金指數日資料、黃金現貨價、美元匯率日報價。本文利用ARJI模型探討四種金融資產的跳躍強度與波動外溢效果,並且以VAR模型做後相關探討。實證結果如下: 1.根據Jarque-Bera常態分配檢定,顯示四項金融變數皆不顯著存在常態分配,可能因隨機的異常事件導致報酬率的變動。 2.在ARJI模型的估計結果顯示,原油、不動產指數、黃金及美元匯率報酬率的跳躍大小的平均數在5%顯著水準下皆為顯著。即表示原油、不動產指數、黃金及美元匯率報酬率均存在著異常資訊所造成瞬時的跳躍行為,其中除美元匯率外,其它變數受到異常資訊所引起的跳躍對於報酬率大多為負面影響。 3.由VAR檢定可發現原油、不動產指數、黃金及美元匯率報酬率的前期皆顯著表示當均衡關係偏離時,原油、不動產指數、黃金及美元匯率需要較大的調整才得以回復均衡。且可看出不動產指數、黃金及美元匯率報酬率的前期係數皆相當大,趨近於1,可以說明這三項變數的前一期變動對於本期影響幾乎為完全正相關。Kaufmann (2004)與Hansen and Lindholt (2004)皆認為,OPEC有能力去影響原油價格。所以推測原油的部分可能是因為中東情勢緊張且戰事頻仍,導致間接影響了西德州原油價格,導致西德州原油前期對於當期的影響不若其它變數為高。

關鍵字

跳躍風險 ARJI模型 REITs VAR

並列摘要


This study discuss the interaction between crude oil price, real estate, gold spot price and U.S. dollar exchange rate since 2005. The sample data is Texas Light Crude Oil Closing Price, U.S. Real Estate Fund Index daily quotation, Gold Spot Price and U.S. Dollar Exchange Rate daily quotation, from June 17, 2005 to September 30, 2011. In this paper, the ARJI model is the tool to explore the jump intensity and volatility spillover effects of these four financial asset, and we use the VAR model to discuss the following issue. The empirical results are as follows: 1. According to the Jarque-Bera normality test, we found that there is no significant evidence to proof the normality distribution of the four financial variables exist, and the fluctuate of the rate of return might be affected by random exception events. 2. ARJI model estimation results show that the mean return of crude oil, real estate index, gold and U.S. dollar exchange are significant at the 5% statistics level. It states that the exceptional information would be the reason of the instantaneous jumping behavior of these four financial elements, and in addition to U.S. dollar exchange rate, the jumping behavior of other variables cause by exceptional information most have negative impact on the return. 3. By the VAR test, we found the T-1 return of crude oil, real estate index, gold and U.S. dollar exchange rate are all significant, indicate that larger amount would be needed for these four financial elements to reinstate the equilibrium relationship when the equilibrium become deviate. And the T-1 return coefficients of real estate index, gold and U.S. dollar exchange rate is quite large, close to 1, indicate that the impacts of the T-1 change of these three variables to the current T term is almost completely positive correlation. For the T-1 coefficient of crude oil, Kaufmann (2004) and Hansen and Lindholt (2004) state that OPEC is powerful able to affect the price of crude oil. Therefore, we speculate that the tension in the Middle East and the war frequency might affect the West Texas crude oil price indirectly, make its impact of T-1 coefficient on the current T is not as strong as the other variables.

並列關鍵字

Jump risk ARJI model REITs VAR model

參考文獻


余佳昇,(2006),油價、金價及英鎊兌美元匯率報酬之共移性與外溢效果,中原 大學國際貿易研究所碩士論文。
詹健宗,(2008),不動產投資信託的影響因素—以美加為例,淡江大學財務金融 學系碩士論文
王怡文,(2007),西德州與布蘭特原油避險策略,淡江大學財務金融學系碩士班 碩士論文。
紀慧君,(2007),原油現貨、期貨與相關性產業之連動關係,淡江大學財務金融 學系碩士班碩士論文。
洪瑞成,(2007),風險值衡量與風險值避險法,淡江大學財務金融學系博士班博 士論文。

延伸閱讀