论文部分内容阅读
针对金融时间序列一般具有非线性、非平稳性、高信噪比和有限样本等特点,将模糊支持向量回归机引入到金融时间序列预测中.设计一种综合模糊隶属度函数,充分考虑到三点:第一噪音会导致错误的回归;第二越靠近预测点的样本对回归的影响越大;第三,离回归线越远的样本,对回归的贡献越大.综合隶属度函数,尽量剔除噪音并给离回归线远的和靠近预测点的样本较大的权值.将采用综合隶属度函数的模糊支持向量回归机应用于羊绒价格序列中,仿真结果表明,本文的基于综合隶属度函数的模糊支持向量回归机在预测精度上有所提高.
Aiming at the characteristics of financial time series, such as nonlinear, non-stationary, high signal-noise ratio and finite samples, the fuzzy support vector regression is introduced into the prediction of financial time series.An integrated fuzzy membership function is designed, Point: the first noise will lead to wrong regression; the second sample that is closer to the prediction point will have greater impact on the regression; third, the sample that is farther away from the regression line will contribute more to the regression. Noise and gives larger weights to the samples that are far away from the tropon and close to the prediction point.The fuzzy support vector regression with integrated membership function is applied to the cashmere price series.The simulation results show that the proposed method is based on the comprehensive membership function Fuzzy support vector regression machine has been improved in prediction accuracy.