The Predition And Influence Factors Analysis Of The International Gold Price

As is well known, the financial market is the artery of all market economic systems. Specially, as a crucial constituent of the financial market, the gold market has been attracting much more interest, by both investors and managers. Since the19th century when the gold market was established, there have been extensive researches on how to predict the gold price. Along with the development of the gold investment and its increasing influence on the market economic system, to understand and further grasp behavior of the gold market evolution, accordingly, becomes important and crucial in stimulating the overall economic development. In this thesis, firstly, possible elements that affect the gold price are discussed. Furthermore, two distinct models are built up. It is found that, the VAR model is more precise in predicting the gold price, thus giving more reliable reference idea for the decision of gold investors and producers. More specifically, the thesis contains the following main contents:(1) By using the theory of the co-integration test and granger causality test, the effect of the following elements on the gold price, including the supply and demand relations, the geopolitics, the dollar index, the oil price, the real interest rate and the inflation, is investigated. Through the co-integration test and granger causality test, it is found that the dollar index and the price of oil are the two key factors affecting the gold price;(2) By using the data form May1987to April2014, the ARMA model and VAR model used to predict the international gold price are built up. Furthermore, based on the models we built up, the evolution of the gold price in the future is predicted;(3) By using the two models, the investigation on predicting the gold price is carried out from aspects of the fitting effect and prediction accuracy. Compared to the outcome from the ARMA model, it is found that the VAR model is more accurate in predicting the gold price, thus can be used for short-term prediction of gold price.