基于神经网络的期铜价格趋势预测研究

Study on the Forecasting of Copper Future Price Trend based on Neural Network

运用X-12-ARIMA季节调整方法,对上海交易所三月期铜月平均价格进行季节性调整,消除了季节因素和不规则因素对铜价的影响。针对季节调整后序列,分别建立了BP、RBF、Elman等神经网络模型,并对期铜价格进行预测。预测效果比较说明,与传统的神经网络相比,Elman神经网络模型具有收敛速度快、预测精度高的特点,能在期铜价格预测方面取得较好的效果。

In this paper, X-12-ARIMA method is used to analyze seasonally fluctuation of month-mean prices of three-month futures copper in the Shanghai Stock Exchange, in order to eliminate the impact of seasonal factors and irregular factors. For the seasonally adjusted series, BP, RBF and Elman neural network models are established, respectively. Then, the three models are used to predict futures copper prices. The predicted results indicate that Elman neural network model has faster convergence speed and higher prediction accuracy than traditional neural networks, and it can achieve better results in the forecasting of futures copper prices.