摘要
影响期货价格短期走势的因素纷繁复杂,具有一定的非线性和随机性,因而难于预测.鉴于神经网络强大的非线性映射能力,利用改进的BP网络,对较难解决的期货价格预测问题进行了研究,提出了一种预测期货价格的方法,并以期货铝的价格为例验证了此方法的有效性.
It is difficult to forecast futures price because the factors which have impact on short-term trend of it are various, complex, nonlinear and stochastic. Considering strong nonlinear mapping ability of neutral network, we apply an improved BP network to forecast futures price and propose a new method in this paper, and then demonstrate the validity of the method by treating forecasting of price of futures aluminum as an example.
出处
《数学的实践与认识》
CSCD
北大核心
2008年第17期49-52,共4页
Mathematics in Practice and Theory
基金
黑龙江省自然科学基金(TA2005-19)
哈尔滨师范大学校基金资助项目(KM2006-11)
关键词
神经网络
BP算法
期货价格
neutral network
BP algorithm
futures price
