| 含缺失值时间序列的ARMA模型拟合 |
| |
| 引用本文： | 张晋昕,王亚拉,何大卫.含缺失值时间序列的ARMA模型拟合[J].中国卫生统计,2000,17(4):197-199. |
| |
| 作者姓名： | 张晋昕 王亚拉 何大卫 |
| |
| 作者单位： | 1. 山西医科大学第二医院,030001
2. 山西医科大学卫生统计教研室 |
| |
| 基金项目： | 本课题为全国统计科学研究计划项目 |
| |
| 摘 要： | 目的 给出一种有效的处理含缺失值时间序列的方法，完成缺失值的内插及ＡＲＭＡ模型的参数估计。方法 用状态空间的Ｍａｒｋｏｖ表达描述时间序列，进而采用Ｋａｌｍａｎ滤波技术。结果 实例分析表明，不仅可以完成缺失值的有效内插，模型拟合效果及预测结虹也甚为柒满意。结论 用基于状态空间表达的Ｋａｌｍａｎ滤波技术，可以实现平稳可逆时间序列中缺失值的内插及ＡＲＭＡ模型拟合。
|
| 关 键 词： | 时间序列 缺失值 ARMA模型 Kalman滤波 |
| 文章编号： | 98028 |
Fitting of ARMA Model to Time Series with Missing Observations |
| |
| Zhang Jinxin,Wang Yala,He Dawei,Deparment of Health Statistics,Shanxi Medical University.Fitting of ARMA Model to Time Series with Missing Observations[J].Chinese Journal of Health Statistics,2000,17(4):197-199. |
| |
| Authors: | Zhang Jinxin Wang Yala He Dawei Deparment of Health Statistics Shanxi Medical University |
| |
| Abstract: | Objective An effective method to deal with time series with missing observations is given.Methods To describe the time series in the form of Markovian representation in the state space,then the Kalman filtering technique is used.Results With the given aproach,interpolation is well accomplished and the goodness of fitting is satisfying.Conclusion Based on the Markovian representation in the state space,Kalman recursive estimation is an effective approach to interpolate the missing values.Simultaneously,the fitting of ARMA models can be fulfilled. |
| |
| Keywords: | Time series Missing observations ARMA models Kalman filter |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
