基于OIVPM的特征值确定ARMA模型的结构
Arma Model Order Determination Based on the Eigenvalues of the Overdetermined Instrumental Variable Produce Moment
-
摘要: 基于最小描述长度(MDL)准则,提出了一种新的自回归滑动平均(ARMA)模型结构 辨识方法.该方法将ARMA模型的结构辨识分两步进行:首先利用超定辅助变量乘积矩 (0IVPM)的最小特征值确定自回归(AR)部分的阶,然后利用协方差矩阵的特征值估计滑 动平均(MA)部分的阶.方法的可行性与有效性通过大量的数值仿真得到验证.Abstract: Based on the MDL criterion, a computationally efficient method for the order determination of ARMA models is proposed in this paper. Order determination is accomplished in two stages, i.e., the AR order is first determined by the minimum eigenvalue of overdetermined instrumental variable product moment (OI VPM ) and the MA order is then estimated by the minimum eigenvalues of a covariance matrix derived from the observed data. The performance of the algorithm is illustrated by extensive numerical examples.
计量
- 文章访问数: 2238
- HTML全文浏览量: 153
- PDF下载量: 1039
- 被引次数: 0