The Linkage Analysis Of Fluctuations Among Gold Price, Oil Price And U.S. Dollar Index

Gold and oil both play important roles during the development of the world economy. In the mean time, as hard currency internationally, the U.S. dollar is the main reserve currency of central banks and shows very significant complementary relationship with gold. In order to study the relationship between gold and oil, as well as the relationship between gold and the U.S. dollar index, this article firstly analyze the trend and the conduction mechanism of the linkage relationship between gold and oil prices and between gold prices and the U.S. dollar index theoretically, followed by analysis from an empirical perspective. This article selects gold prices, oil price and U.S. dollar index from January2,1986to June14,2013as the sample data. We at first establish VAR model, Johanson cointegration test and Granger causality test based on VAR model. The VAR model has a relatively higher fitting degree, and the model results indicate that the gold price has significantly positive correlation with the oil prices lagged one period, while show significantly negative correlation with the U.S. dollar index lagged one period in the same time. Besides, the same period correlation matrix shows that gold prices, oil prices and the U.S. dollar index do not affect each other obviously in the same period. The results of Johanson cointegration test show that there is a long-term equilibrium relationship between gold prices, oil prices and the U.S. dollar index. While Granger causality test results show that both the oil prices and the U.S. dollar index are the Granger reasons of gold prices, and the impact from oil prices is one-way, while the influence between the U.S. dollar index and gold prices is two-way. Next the article separately establish the model between the gold and the U.S. dollar index market as well as the model between the gold and oil markets based on the DCC-GARCH model. The DCC-GARCH model has two stages. The first stage of the modeling process is to build single variable GARCH model for gold, oil and the U.S. dollar index return series, the results show that fluctuations in the gold, oil and the U.S. dollar index yield markets is significantly continuous. In terms of the fluctuation trends, the gold and oil markets’volatility shows a certain degree of consistency on trends, sometimes there appear large peaks simultaneously. On the other hand, the U.S. dollar index displays more frequent fluctuations, which is similar to the shock state. There also appear peaks in few cases with the gold and the oil market simultaneously. The second stage separately build bivariate DCC-GARCH model between the gold and the U.S. dollar index markets as well as between the gold and the oil markets, the results show that the degree of continuous correlation between the gold market and the other two markets is relatively higher. Moreover, the mean value of dynamic conditional correlation coefficients are significantly different from zero, the nature of positive or negative is also in line with expectations, yet the mean value of the correlation coefficients is obviously lower than1, indicating a weaker ability to avoid the risk of each other between the two markets. From dynamic conditional correlation coefficients charts, we can see that the correlation between the gold market and the oil market is more regular, basically bouncing fluctuate within a certain range, and the linkage between the two markets is in a weak rising channel in recent years. While the correlation between the gold market and the U.S. dollar index market performs large fluctuations, especially that the linkage between the two markets is comparatively weaker from1991to2000, which is relevant with the fact that the United States flatter the virtual economy and manufacture the dot com bubble.