基于EMD与BP神经网络的中国股票指数期货价格预测

应用经验模态分解算法（EMD）和BP神经网络理论提出了我国股指期货市场价格走势预测模型。首先应用EMD分解算法把股指期货价格序列分解成不同尺度的内禀模态分量（IMF）,再通过重复试验的方法运用BP神经网络对股指期货价格序列和分解得到的所有IMF的数据序列进行训练,得到股指期货价格的预测模型,并对股指期货价格进行预测。实验表明,通过该方法得到的预测值与股指期货的实际价格有着很高的拟合度。

Prediction of Stock Index Future Price Based on EMD and BP Neural Network

The prediction model of the price movements of stock index futures in the market of China based on empirical mode decomposition(EMD) algorithm and BP neural network theory is proposed.By using the EMD algorithm,the price series of stock index futures is decomposed into Intrinsic Modal Functions(IMF)of different scales.The price series and the data sequence of IMF are trained by the BP neural network.The prediction model is adjusted via repeated experiments to predict the prices.Simulation results show that the predicted values are in agreement with the actual value of stock index futures.