为有效提高神经网络集成的泛化能力,先利用量子粒子群和主成分分析提高集成个体的泛化能力,再利用泛化能力强的支持向量机回归集成生成输出结论,建立一个基于支持向量机的粒子群神经网络集成股市预测模型.试验表明,该模型能有效提高神经网络集成系统的泛化能力,预测精度高,稳定性好. In order to effectively improve the generalization ability of neural network ensemble, quantum particle swarm optimization and principal component analysis are used firstly to improve the generalization ability of integrated individuals, and then the support vector machine regression with generalization ability is used to generate integrated output conclusions. Then a support vector machine Particle swarm optimization neural network integrated stock market prediction model.The experiment shows that this model can effectively improve the generalization ability of the neural network integrated system with high prediction accuracy and good stability.