摘要
应用BP神经网络理论提出了我国股指期货市场价格走势短期预测模型。首先根据实验数据的特点分别构建单因素、多因素BP神经网络预测模型,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进行训练,从而对股指期货价格进行预测。结果表明,通过BP神经网络预测模型得到的预测值与股指期货的实际价格有着很高的拟合度。
The prediction model of the price about the short-time movements of stock index futures was proposed with BP neural network theory in the market of China. The single-factor and multi-factors BP neural networks is built according to the characteristics of the raw datas. The prediction model is adjusted through repeated experiments to predict the prices. Simulation results show that the predicted values are a greement with the actual value of stock index futures.
出处
《青岛大学学报(自然科学版)》
CAS
2012年第3期93-96,共4页
Journal of Qingdao University(Natural Science Edition)
基金
国家自然科学基金资助项目(709710071)
关键词
BP神经网络
股指期货
价格预测
BP neural network
stock index tuture
prediction of price