| 股票价格短期预测的LM遗传神经网络算法 |
| |
| 引用本文： | 肖菁,潘中亮. 股票价格短期预测的LM遗传神经网络算法[J]. 计算机应用, 2012, 32(Z1): 144-146,150 |
| |
| 作者姓名： | 肖菁 潘中亮 |
| |
| 作者单位： | 1. 华南师范大学计算机学院,广州,510631
2. 中山大学计算机科学系,广州,510006 |
| |
| 基金项目： | 教育部博士点新教师基金资助项目 |
| |
| 摘 要： | 针对传统神经网络在股票价格预测中易陷入局部最优和预测精度偏低的问题,提出了一种改进的神经网络算法进行求解.改进的算法基于LM算法建立了改进的三层BP神经网络对股票价格建立预测模型,避免算法陷入局部最优,并运用遗传算法优化神经网络的权值和阈值,提高神经网络对股票价格的预测精度.采用Matlab对上述算法模型进行仿真测试,通过预测4支股票价格的实验证明改进后的神经网络预测算法对股票价格的短期预测有较好的精度和稳定性.
|
| 关 键 词： | LM算法 遗传算法 BP神经网络 股票价格短期预测 葛南维尔移动平均线 |
Stock price short-time prediction based on GA-LM-BP neural network |
| |
| XIAO Jing , PAN Zhong-liang. Stock price short-time prediction based on GA-LM-BP neural network[J]. Journal of Computer Applications, 2012, 32(Z1): 144-146,150 |
| |
| Authors: | XIAO Jing PAN Zhong-liang |
| |
| Affiliation: | 1.School of Computer Science,South China Normal University,Guangzhou Guangdong 510631,China; 2.Department of Computer Science,Sun Yat-sen University,Guangzhou Guangdong 510006,China) |
| |
| Abstract: | In order to deal with the problems of local optima and low predicting accuracy in stock price prediction using traditional neural networks,Levenberg-Marquardt(LM) algorithm is introduced to improve the three-layer back propagation(BP) neural networks in this paper to prevent the local optima.Genetic algorithm(GA) is used to optimize the weight values and thresholds in BP neural networks to obtain high accuracy of predicting the stock price.The proposed method is simulated in Matlab and the experimental results of predicting price of 4 stocks show the proposed algorithm is effective and promising. |
| |
| Keywords: | Levenberg-Marquardt algorithm Genetic Algorithm(GA) BP neural network stock price short-time prediction Granvile moving average |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
