【摘要】 为了降低股价趋势中所含的噪声信息和输入变量的相关性对股价预测的影响，构造信噪比（SNR）特征向量，从而提出基于信噪比的KPCA-SVM-KNN的股价预测模型，并借用国内A股市场的股票价格相关数据进行实证分析，结果表明：1)SNR特征向量的加入可以提高股票分类准确率，进而增加股价预测精度；2）与现有的SVM-KNN算法相比，所提出的KPCA-SVM-KNN算法可以提高股价预测准确度，减少预测误差，从而为决策者的投资决策提供帮助。更多还原

【Abstract】 In order to reduce the influence of the correlation between the noise information and the input variables in the stock price trend on the stock price prediction,the paper constructs the signal-to-noise ratio(SNR)eigenvector,and then proposes the KPCA-SVM-KNN stock price prediction model based on the SNR,and makes an empirical analysis using the stock price related data of the domestic A-share market. The results show that compared with the existing SVM-KNN algorithm,the KPCA-SVM-KNN algorithm can improve the accuracy of stock price prediction and reduce the prediction error,so as to provide help for decision makers in investment decision-making.更多还原