引用本文:徐南荣 ,陈为余.利用相关熵估计ARMA模型的参数[J].控制理论与应用,1987,4(4):34~40.[点击复制]
Xu Nanrong, Chen Weiyu.PARAMETER ESTIMATION FOR ARMA MODEL USING CORRELATION ENTROPY[J].Control Theory and Technology,1987,4(4):34~40.[点击复制]
利用相关熵估计ARMA模型的参数
PARAMETER ESTIMATION FOR ARMA MODEL USING CORRELATION ENTROPY
摘要点击 1241  全文点击 494  投稿时间:1985-12-10  修订日期:1986-09-28
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  
  1987,4(4):34-40
中文关键词  
英文关键词  
基金项目  
作者单位
徐南荣 ,陈为余 南京工学院 
中文摘要
      本文提出估计ARMA模型参数的一种线性方法。根据信息熵的概念,利用相关熵作为衡量残差序列与白噪声序列逼近程度的一个定量指标。借助于相关熵的变化规律来确定拟合观测序列的AR模型的阶次。估计出ARMA模型参数的初值,并运用简单的线性方法实现ARMA模型参数的精估计。数字仿真结果说明此方法优于其他类似的线性估计方法。
英文摘要
      In this paper a linear method for parameter estimation of ARMA model is proposed. According to the concepts of information entropy, the correlation entropy is used as a quantitative performance index to measure the degree of closeness between the sequences of residues and white noise. The order of AR model fitted to observed data is selected according to the variation of the correlation entropy, and the initial values of the ARMA model parameters are estimated. These parameters are eventually identified by using simple linear method. Digital simulation for some examples shows that the method proposed is superior to other similar ones.