机构地区: 内蒙古民族大学理工学院
出 处: 《吉林大学学报(信息科学版)》 2002年第3期68-72,共5页
摘 要: 考虑到投资者的主要目的是盈利这一重要因素 ,提出了一种改进的径向基函数 (RBF:Radial basis func-tion)神经网络方法。在 RBF网络的误差函数中增加了利润、时间和趋势信息 ,并采用基于梯度下降的误差纠正算法对网络进行训练。对股市综合指数进行预测的结果表明 ,该方法在提高投资收益的意义下 ,提高了神经网络模型在金融领域的预测性能。 Considering the important factor that the main purpose for the investor is to make high profit,a modified RBF(Radial basis function) neural network method is introduced. The information of the profit, time and trend is added in the error function of RBF neural network. The algorithm of error correction based on the gradient descent is used to train the neural network. The results of predicting the share index show that the modified method raises the predicting characteristics in financial field using neural network models in respect to raising the profit of the investment.
领 域: [自动化与计算机技术] [自动化与计算机技术]