石油和黄金是推动当今社会稳定发展的必备商品，对社会各方面的发展意义重大。且由于经济的迅速发展，经济市场的波动带来的联动性以及潜在的因果关系也日益密切，该因果关系在大宗商品市场异常显著。因此，石油和黄金在价格上的波动关系以及因此产生的风险和利润引起了国内外学者的广泛关注。尤其在近现代，石油与黄金它们在价格上的走势基本一致，因此石油与黄金它们在价格以及收益率方面有没有固定的联动关系，有没有因果关系，值得去探讨和研究。且Granger因果关系自1969年提出以来便经常用来探讨经济学相关问题，这种研究方法已经很成熟，故采取Granger因果关系检验方法来探讨石油和黄金之间的关联是可行的。 本文首先详细阐述了线性、非线性Granger因果关系的理论知识、检验步骤，以及与其相关的平稳性检验、协整检验、VAR模型的建立等理论知识。然后基于国内外一些相关的研究文献，探讨了黄金与石油它们在价格上存在的联动关系，得到了石油与黄金价格的影响因素、历史走势以及它们联动的原因。最后选取了1987年5月到2018年2月每月最后一个交易日的BRENT（布伦特原油）价格数据，以及从世界黄金协会官网中选取的相同时间段的黄金价格数据，再利用黄金、石油的价格数据计算得到它们的收益率数据，用这两组数据对石油和黄金之间的关系进行实证分析，本文主要研究内容和研究发现如下： 1.利用价格数据做了平稳性检验、EG协整检验和线性Granger因果检验。通过协整检验发现黄金与石油它们的价格数据是协整的，有着长期的均衡关系，且它们在价格上有正相关性，油价的波动定会引起金价的波动。同时黄金和石油它们在价格上有着显著的双向线性Granger因果关系。 2.利用价格数据计算得到石油与黄金收益率的数据，对其做了平稳性检验、进行了VAR建模以及线性、非线性Granger因果检验。得到石油与黄金的收益率数据之间通常情况下没有线性Granger因果关系的结论。但基于HJ（Hiemstra-Jones）检验得到了由石油收益率到黄金收益率的显著的单向非线性Granger因果关系的结论。

Oil and gold are essential commodities to promote the stable development of today's society, and are of great significance to the development of all aspects of society. And because of the rapid economic development, the linkages and the potential causality brought about by the volatility of the economic market are becoming increasingly close. This causal relationship is unusually significant in the commodity market. The volatility of oil and gold in terms of price and the risks and profits generated by it have attracted widespread attention from scholars at home and abroad. Especially in modern times, oil and gold have basically the same price trend. Therefore, whether oil and gold have a fixed linkage mechanism and causal relationship in terms of price and profitability, it is worth exploring and researching.. And Granger causality has been used to explore economics-related issues since its introduction in 1969. This research method is very mature. Therefore, it is feasible to use the Granger causality test method to explore the relationship between oil and gold. In this paper, we first elaborate the theoretical knowledge, test steps of linear Granger causality and nonlinear Granger causality, and its related stationary test, cointegration test, and establishment of VAR model. Then based on some relevant research literature at home and abroad, discussing the relationship between gold and oil in price. Get the factors affecting the price of oil and gold, historical trends and the reasons for their linkage. Finally, the BRENT price data of the last trading day of May 1987 to February 2018 was selected, and the gold price data of the same time period selected from the World Gold Council. Reuse the price data of gold and oil to calculate their yield data, and use these two sets of data to empirically analyze the relationship between oil and gold. The main research contents and research findings of this paper are as follows: 1.Use the price data to do the stationarity test, EG cointegration test and linear Granger causality test. Through the cointegration test, it is found that the price data of gold and oil are cointegrated and have a long-term equilibrium relationship. And they have a positive correlation in price, and fluctuations in oil prices will cause fluctuations in gold prices. At the same time, gold and oil have significant two-way linear Granger causality in price. 2.Use price data to calculate data on oil and gold yields. It was tested for stationarity, VAR modeling, and linear, nonlinear Granger causality tests. There is usually no linear Granger causal relationship between the oil and gold yield data, but based on the HJ test, the conclusion of the one-way nonlinear Granger causality from oil yield to gold yield is obtained.