本研究內容包含二個議題,首先利用Ling and McAleer(2003)提出之VARMA-GARCH模型,分別探討高息貨幣-澳幣、紐幣及南非幣與近月黃金期貨彼此間波動外溢現象。再者將三種貨幣各別搭配黃金期貨,透過計量模型分別計算出最適投資組合權重與最適避險比例,比較全部持有外幣資產之未避險投資組合(PF I) 及搭配近月黃金期貨後之已避險投資組合 (PF II)之避險效果。 在波動外溢現象方面,本研究發現在全樣本期間中,前期黃金市場的波動會影響外匯市場之波動,存在單向波動外溢,但在實施貨幣量化寬鬆(QE1)前,其波動外溢現象較不明顯,反而在實施貨幣量化寬鬆(QE1)後,預期美元逐步貶值,投資人提高持有黃金保值的意願,澳幣、紐幣及南非幣匯率與黃金彼此間存在雙向波動外溢,其中以南非幣承受的波動程度最高。 在避險效果方面,本研究發現,在全樣本期間及實施貨幣量化寬鬆(QE1)前、後,採用最適投資組合權重(OPW)所建構之投資組合,其避險效果表現優於最適避險比例(OHR)所建構之投資組合,無論是利用報酬的變異數變動程度(HEV)衡量,或是衡量每單位風險所能創造之報酬(HER)。再者,無論是利用最適投資組合權重(OPW)或是最適避險比例(OHR)建構之已避險投資組合,在實施貨幣量化寬鬆(QE1)後之避險效果多優於實施貨幣量化寬鬆(QE1)前,顯示當美國實施貨幣量化寬鬆(QE1)後,市場投資人將黃金納入投資組合中,確實能達到避險需求。 最後,本研究透過多種不同計量模型來衡量投資組合之避險效果,發現不管是全樣本期間亦或是實施貨幣量化寬鬆(QE1)前、後,以Bollerslev(1990)之CCC-GARCH模型及Engle(2002)之DCC-GARCH模型衡量之已避險投資組合,其避險效果相對較佳,特別是在澳幣及紐幣與黃金期貨之投資組合上。
In this article, we take the bivariate VAR(1)-GARCH(1,1) model of Ling and McAleer (2003) to examine the extent of volatility spillover for each pair of the exchange rate of high yield currencies and nearby gold futures. We also analyze the optimal portfolio weight (OPW) and hedge ratio (HER) for the exchange rate-gold futures portfolio holdings with respect to the results. Regarding the extent of the volatility transmission, we find that the foreign exchange market is significantly affected by gold´s past volatility in the whole period, but on the contrary, it is not. Before the United States executed the first quantitative easing (QE1), the effects of volatility spillover are not obvious. Many investors expect that the dollars is going to depreciate after executing QE1, consequently, there is a strong evidence to suggest that there are cross effects of conditional volatility between the foreign exchange and gold futures, especially in South African Rand (ZAR) against gold futures. Another topic we analyze is the hedging effectiveness, we find no matter what indicators (i.e. HEV or HER) we estimate, the hedging effectiveness for the portfolio we create with the OPW is better than the one with the HER in the whole period, ex ante QE1 and ex post QE1. In addition, the hedging effectiveness in most hedged portfolios, except for ZAR-gold futures portfolio, are better in ex post QE1 than in ex ante QE1. The empirical results suggest that the hedged portfolio including gold futures is a preferable choice to investors when the United States executed quantitative easing. Our back-testing procedures, finally, suggest that taking the CCC-GARCH of Bollerslev (1990) and the DCC-GARCH of Engle (2002) often lead to diversification benefits and hedging effectiveness better than those of commonly used multivariate GARCH models, especially in Australian Dollar (AUD)-gold futures and New Zealand Dollar (NZD)-gold futures portfolios.