Research On The Mineral Product Price Forecasting Model And Its Application

The trend of minerals price significant has great impact on production management of mining enterprises. In the determination of industrial index of mine production, development of mine production plan, mineral sales plan, mineral prices forecast can provide useful reference information to help mining enterprises to accurately complete production management decisions. Based on gold prices as an example, the gold, antimony, tungsten minerals price forecasting models were studied and analyzed, and demonstrated the scientific and practical of analysis of the mineral price forecasting model.Regression analysis forecasting model, gray GM(1,1) forecasting model, exponential smoothing forecasting model and ARIMA forecasting model four single forecast models were used to analysis and forecast the mineral prices. Due to the limitation of Single forecasting model, the minimum sum of squared errors combined forecasting model, the minimum absolute errors combined forecasting model and Back-Propagation neural network ensemble forecasting model were imported for further analysis and forecast of mineral prices. Then calculated the errors of each minerals price forecasting model, prediction fitting results with precise index and judged the validity of the model for each forecasting model analysis and evaluation. Finally, analyzed the price forecasting model application in cutoff grade and other areas of mine production management. The main works of this paper are as follows.1. Analyzed the knowledge of mineral prices and factors of fluctuations. Based on gold prices as an example, Regression analysis and forecasting model, gray GM (1,1) prediction model, exponential smoothing prediction model and ARIMA forecasting model four single forecast models were used to establish the gold price forecasting model for forecasting gold price.2. Established the linear optimal combined forecasting model named Minimum error sum of squares combination forecasting model and minimum absolute errors combined forecasting model on the basis of the gold forecasting model and Back-Propagation neural network combined forecasting model belongs to non-linear combination forecast model. Summarized each individual prediction model for the same forecast of gold prices in the coming year.3. Used the precision index for analyzing the fitting results of prediction model of each mineral price, verified the validity and applicability of price prediction model and the combination forecast model. And proved that the combination forecasting model prediction accuracy was higher than any single forecast model in mineral price forecast.4. Established the gold, antimony, tungsten price forecasting model for Xiangxi Gold Mine of Chenzhou Mining and provide the effective and practical information for mine production management decisions. And the introduction of mineral combination forecasting provided a new method for the calculation of the mine cutoff grade.