【摘要】 研究准确优化预测股票价格问题,针对影响股票价格具有非线性、不稳定的特征,股票价格由于受到社会经济因素的影响,变化大。采用传统神经网络方法在股票价格预测中易陷入局部极小值,泛化能力受到影响。为了提高股票价格精度,提出一种基于粒子群优化算法(PSO)的RBF神经网络(RBFNN)股票价格预测模型。利用粒子群优化算法的良好的寻优能力,对RBF神经网络参数进行优化,从而加快RBF神经网络运算速度,并提高了RBF神经网络的预测精度。利用粒子群优化的RBF神经模型对上证指数(000001)股票价格进行了验证性测试和分析,实验结果表明,相对于各参比模型,经过粒子群优化的RBF神经网络模型预测方法有更好的收敛性,更强的学习能力,显著地提高了预测精度,可为预测提供依据。更多还原

【Abstract】 Influence factors of the stock price have high nonlinear characteristics,and cause the neural network stock price forecasting easily trapping into local minimum values.In order to improve the accuracy of stock price,a stock price prediction model was proposed based on the particle swarm optimization algorithm(PSO) and RBF neural network(RBFNN).Firstly,we used particle swarm optimization algorithm to optimize the parameters of the RBF neural network for its good optimization ability,thus accelerated the computing speed of the RBF neural network and improved the prediction accuracy of the RBF neural network.The proposed algorithm was tested and analyzed by Shanghai index(000 001) stock prices.Experimental results show that,compared with the reference model,the proposed prediction method has better convergence and strong learning ability,and significantly improve the accuracy of the predictions.更多还原