Font Size: a A A

Dynamic Prediction And Influence Factors Analysis On The Price Of Gold

Posted on:2017-05-09Degree:MasterType:Thesis
Country:ChinaCandidate:L LiuFull Text:PDF
GTID:2309330485974824Subject:Applied statistics
Abstract/Summary:PDF Full Text Request
Gold is one kind of product with financial attributes, and it is one of the important spot and stock contract products in the commodity market. In the world economic crisis in 2008, a variety of financial products have showed weak in the present situation. But only the gold market has outstanding performance. Recently the stock market of C hinese has a substantial fluctuations, but the price of gold has small fluctuation. Based on the data analysis method to explore the rule of gold price’s change has positive significance, whatever for Investors、producers or government regulators.The structure of this article is divided into five parts and studied from different aspects. On the basis of the research for writing background、research ideas、content and method. First, for the yield on the gold price, the article has carried on the testing using the model of cointegration outlier detection、boxplot and so on. And found out the date corresponding to the abnormal points. Secondly, Build the model of ARIMA-GARCH, respectively, using removed and unremoved abnormal point data. The results showed that the model with the removed data has a higher significance. Then we have used the model to analyze the volatility of gold price and carried on a forecast. In addition, the article using support vector machine(SVM) predicted the price of gold in the technology of data mining. Finally we compared the two different results and found that the model of svm has a better prediction effection. Thirdly, on the basis of the analysis for gold price’s volatility, we found it is meaningful that to study the factors which influencing the price of gold. We studied the previous literature and summarized some influencing factors. Then we recommend two new factors with users and media index. In follow-up studies, we has build different models of VAR with the different interpretation variables. The study found that the overall effect of model became more better after we recommend new factors. In addition, we also analyzed the different model respectively with monthly and daily data, and then put forward the results of comparison. The last, the conclusions and recommendations of this article was presented.Through the oretical and empirical analysis and all sorts of comparative analysis, we have carried on the comprehensive analysis from aspects of the gold price. By studying the literature found that the study of the price of gold has been carried out from many aspects. Such as volatility, influence factor analysis and so on. But this article is the analysis of China’s gold price. So because of the particularity of China’s national conditions, the analysis and conclusion of this paper are still need further perfect.
Keywords/Search Tags:The Price of Gold, Volatility, Anomaly Detection, Support Vector Machines, Vector Auto-regressive Model
PDF Full Text Request
Related items