基于径向基函数网络的改进算法的股市数据预测

Prediction of Stock Market Data Based on Improved Algorithm of Radial Basis Function Network

径向基函数网络的性能在很大程度上取决于径基函数中心位置的选取。本文通过结合输入输出矢量从而得到扩展矢量的方式改进了常用的确定径基函数中心的HCM算法。股市数据预测的实验结果表明 :改进的HCM算法的网络的性能有了明显的改善。

Performance of radial basis function network highly depends on the locations of radial basis function centers. A improved Hard C-Means algorithm which is often used to confirm the locations of centers is put forward in the paper by concatenating the output vector to the input vector to gain a augmented vector. The experiment result of prediction of stock market data indicates that the network performance of improved Hard C-Means algorithm is ameliorated obviously.