作　　者： ; ;

机构地区： 广东松山职业技术学院

出　　处： 《电气传动自动化》 2009年第6期17-19,42,共4页

摘　　要： 介绍了神经网络的基本概念,BP算法是前馈神经网络中最终要的算法模型,但由于BP算法收敛速度慢,为了加速算法收敛,增加神经网络的稳定性,提出了L-M对股票指数预测模型。借助神经网络对非线性函数的逼近能力,对上证综合指数股价进行单步预测。用L-M和BP对预测结果进行比较,证明L-MBP算法用于指标预测准确性更高。 For feed-forward neural network,BP algorithm is among the most important neural network algorithms.BP algorithm has its local minima and its slow training speed.In order to speed the convergence of BP algorithm and improve the network stability,a Levenberg-Marquardt algorithm based on optimal theory is put forward.By its nonlinear disposal capacity and good stability,the "single pace" stock price of Shanghai stock market is also predicted.The final result of prediction is compared by using BP and L-MBP.

关 键 词： 神经网络 预测 股票价格

领　　域： [经济管理]