摘要
研究股票价格预测问题,股票价格变化具有非线性、时变性,传统线性预测模型难以准确刻画股价变化规律,且非线性神经网络存在过拟合、局部最小等缺陷,预测精度比较低。为提高股票价格预测精度,提出一种基于粒子群优化支持向量机的股票价格预测模型。利用粒子群算法良好的寻优能力,对支持向量机参数进行优化,加快支持向量机学习速度,再采用非线性预测能力优异的支持向量机对股票价格进行预测。以南天信息股票价格对模型性能进行仿真,实验结果证明,支持向量机预测模型能全面反映股票价格变化的非线性的时变规律,获得更高预测精度,预测结果对股民实际操作具有较大的指导价值。
As the stock become an important part of people＇s economic life,the problem of stock price forecasting has become a significant concern in recent years.The stock price is non-linear time series data.The fact is that the traditional forecasting method lacks of precision and adequate preparation for stockholders.Therefore,this paper tried to make use of support vector（SVR） regression to establish a model of stock price forecasting for the opening price on the third day.Meanwhile,in order to get the better forecasting model,various parameters were optimized through particle swarm optimization（PSO） method.The results of simulation experiments show that the model can quite exactly forecast the opening prices on the third day,so it has larger guidance value to stockholders.
出处
《计算机仿真》
CSCD
北大核心
2012年第4期379-382,共4页
Computer Simulation
关键词
支持向量回归机
股价预测
粒子群优化算法
Support vector regression（SVR）
Stock price forecasting
Particle swarm optimization method
