| 基于模糊SVM与周期分析的股票预测 |
| |
| 引用本文： | 李天生,李霆.基于模糊SVM与周期分析的股票预测[J].五邑大学学报(自然科学版),2008,22(3):41-45. |
| |
| 作者姓名： | 李天生 李霆 |
| |
| 作者单位： | 五邑大学,信息学院,广东,江门,529020 |
| |
| 摘 要： | 基于模糊SVM与周期分析的股票预测方法通过选取曲率变化大、形式简单的幂函数作为候选隶属度函数，并采用格子搜索法寻找最优参数，确定出最优模糊隶属度，同时又结合寻找股票波动的工具：市场周期分析法．仿真实验表明：在利用模糊SVM训练时间序列数据集时，该方法比目前常用的选择模糊隶属度、以及单独用最优模糊隶属度方法的效果都好．
|
| 关 键 词： | 模糊支持向量机 最优模糊隶属度 股票波动 股票预测 |
Stock Forecasts Based on Fuzzy SVM and Cycle Analysis |
| |
| LI Tian-sheng,LI Ting.Stock Forecasts Based on Fuzzy SVM and Cycle Analysis[J].Journal of Wuyi University(Natural Science Edition),2008,22(3):41-45. |
| |
| Authors: | LI Tian-sheng LI Ting |
| |
| Institution: | ( Information School, Wuyi University, Jiangmen 529020, China ) |
| |
| Abstract: | A new method of stock forecasting based on Fuzzy Support Vector Machine （SVM） and cycle analysis is advanced. Optimal fuzzy membership is determined by choosing power function as candidate membership function and optimal fuzzy membership is determined by adopting a grid search method, Also a tool for searching stock fluctuation is used： market cycles analysis, The simulation results show that the method has better performance than the usual fuzzy membership selection method and the single use of the optimal fuzzy membership method in the training of time series data set by using Fuzzy SVM. |
| |
| Keywords: | Fuzzy Support Vector Machine optimal fuzzy membership stock fluctuation stock forecasting |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
