The Research On The Time Series Analysis Method Application In The Trend Prediction Of The Gold Price

Gold is not only a precious metal but a hard c urrency. Investing in gold cannot guarantee the investor’s assets eroded by inflation, but also easy to like stocks and real estate investments that would face the market crash, so deeply loved by the people. Historical price data using data mining for gold can effectively predict the gold price movements, guiding investment decisions of investors with a wide range of application value.Aiming at the forecasting applications of the trend of gold price, the time-series features deep in the price of gold is analyzed, and besides the common nonlinear, non-stationary and dynamic features, it also contains high noise and non-normal characteristics are summarized. The present situation on which the domestic and foreign researchers about the price of gold are analyzing is introduced, and then the knowledge of time series and statistical software for quantitative interpretation are also used. The innovative applications of the gold price forecast in this paper mainly in the following three aspects:1) Bases on the principle of gray model, aiming at high noise, non- normality and other characters of the trend of the gold price, a method of forecasting the gold price trend bases on gray model is proposed and applied to the gold price forecast by actual data validation the effectiveness of this method.2) Bases on the principle of differential autoregressive moving average model for the inherent characteristics of the gold price movements, a differential autoregressive moving average model prediction bases on gold price movements is proposed, and applied to predict the price of gold fields, by the actual data demonstrate the effectiveness of this method.3) Bases on the real price of gold, the weighted moving average, exponential smoothing, and ARIMA models and other models gray gold price trend forecasting accuracy and effectiveness are analyzed comparatively and the advantages and disadvantages of these four models and apply scene are summarized.The forecast of the trend of the gold price base s on gray model, the fo recast of the trend of the gold price base s on ARIMA model, moving average model and exponential smoothing model are used to analyze in this paper, and the predict results of the model are analyzed for comparison. By the comparative analysis of the fitting effect of the four models, they all can describe the trend of the gold price, especially autoregressive moving average model, whose fitting effect was best. This model shows a strong practical value in the short-term forecast by empirical research. Based on this and other crucial technology research and breakthrough, finally this paper realizes the application of the whole system.