|
|||||
|
|
| 非平稳时间序列分析的WAVELET—改进GM(1,1)组合方法及其应用 | |
| 引用本文： | 张贵杰,任永泰,王福林,付强.非平稳时间序列分析的WAVELET—改进GM(1,1)组合方法及其应用[J].数学的实践与认识,2012,42(1):135-140. |
| 作者姓名： | 张贵杰 任永泰 王福林 付强 |
| 作者单位： | 1. 东北农业大学工程学院,黑龙江哈尔滨,150030 2. 东北农业大学理学院,黑龙江哈尔滨,150030 3. 东北农业大学水利与建筑学院,黑龙江哈尔滨,150030 |
| 基金项目： | 黑龙江省青年基金“哈尔滨市及城近郊区地下水水位水质预警研究”(QC08C64) |
| 摘 要： | 提出非平稳时间序列分析的WAVELET—改进GM(1,1)组合方法.首先利用Mallat算法对非平稳时间序列进行小波分解;然后采用能量阈值选择策略对高频系数进行处理,并将其与低频系数进行小波重构;最后运用改进的GM(1,1)方法预测.该方法不仅能充分拟合低频信息,而且可避免对高频信息的过拟合.实验结果证明,该方法比传统的非平稳时间序列预测方法具有更高的预测精度. |
| 关 键 词： | 小波分解 非平稳时间序列 能量阈值 改进GM(1 1)模型 预测 |
Wavelet-improved GM(1,1) Method in the Non-stationary Time Series and Its Application |
|
| ZHANG Gui-jie , REN Yong-tai , WANG Fu-lin , FU Qiang.Wavelet-improved GM(1,1) Method in the Non-stationary Time Series and Its Application[J].Mathematics in Practice and Theory,2012,42(1):135-140. | |
| Authors: | ZHANG Gui-jie REN Yong-tai WANG Fu-lin FU Qiang |
| Institution: | 1.College of Engineering,Northeast Agricultural University,Harbin 150030,China) (2.College of Science,Northeast Agricultural University,Harbin 150030,China) (3.College of Water Conservancy and Building Engineering,Northeast Agricultural University,Harbin 150030, China) |
| Abstract: | In this paper,WAVELET—improved GM(1,1) Method is proposed in the nonstationary time serious.By wavelet decomposition,the non-stationary time series were decomposed. Then the energy threshold selection strategy was adopted to process the high frequency coefficients.By wavelet transform,make the processed high frequency coefficients and low-frequency coefficients reconstruct.Use the improved GM(1,1) method to make forecast. This new method avoids the over-fitted for high frequency signals,and adequately fits the low singal of the non-stationary time series,so better predicting performance can be obtained. Experiments show the novel method is higher accuracy in comparison with the traditional one. |
| Keywords: | wavelet decomposition non-stationary time series energy threshold improved GM(1 1) model prediction |
| 本文献已被 CNKI 万方数据 等数据库收录！ | |