| 引用本文： |
-
汪灵枝,罗朝晖,韦增欣,赵秋梅.量子粒子群优化神经网络集成股市预测模型研究[J].广西科学,2010,17(4):324-327. [点击复制]
- WANG Ling-zhi,LUO Chao-hui,WEI Zeng-xin,ZHAO Qiu-mei.A Neural Network Ensemble Forecasting Model Research of Stock Market Based on Quantum Particle Swarm Optimization[J].Guangxi Sciences,2010,17(4):324-327. [点击复制]
|
|
| 摘要: |
| 利用量子粒子群优化神经网络集成个体的网络结构和连接权值,对集成个体进行支持向量机回归集成,建立一个新的量子粒子群优化神经网络集成股市预测模型。新模型能有效提高神经网络集成系统的泛化能力,易操作,稳定性好,预测精度高,具有良好的应用前景。 |
| 关键词: 优化 股市预测 量子粒子群 支持向量机 神经网络 集成 |
| DOI： |
| 投稿时间：2010-03-11 |
| 基金项目: |
|
| A Neural Network Ensemble Forecasting Model Research of Stock Market Based on Quantum Particle Swarm Optimization |
|
WANG Ling-zhi1,2, LUO Chao-hui3, WEI Zeng-xin2, ZHAO Qiu-mei2
|
| (1.Department of Mathematics and Computer Science, Liuzhou Teachers College, Liuzhou, 545004, China;2.College of Mathematics and Information Science, Guangxi University, Nanning, Guangxi, 530004, China;3.Department of Mathematics and Computer Science, Baise University, Baise, Guangxi, 533000, China) |
| Abstract: |
| A novel neural network ensemble forecasting model based on quantum particle swarm optimization (QPSO) was proposed. The QPSO algorithm is used to evolve neural network architecture and connection weights, to generate different individual of neural network. Then Support Vector Machine is used for regression ensemble. Empirical results reveal that the prediction is generalization ability.The illustration and testing reveal that the ensemble model proposed can be used as an alternative forecasting tool for stock market forecasting in achieving greater accuracy and improving prediction quality further. |
| Key words: optimization stock market forecast quantum behaved particle swarm optimization support vector machine neural network ensembles |
