Bootstrap-based estimates of uncertainty in subspace identification methods

被引:35
作者
Bittanti, S [1 ]
Lovera, M [1 ]
机构
[1] Politecn Milan, Dipartimento Elettron & Informaz, I-20133 Milan, Italy
关键词
identification algorithms; subspace methods; statistical analysis; uncertainty;
D O I
10.1016/S0005-1098(00)00081-9
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The problem of evaluating model uncertainty in the framework of subspace identification methods is considered and the use of the methods of computational statistics is proposed. In particular, a procedure for the estimation of the standard error of the eigenvalues and of the frequency response of the identified models is presented. Simulation results demonstrate that accurate estimates of the standard error can be obtained by this approach. (C) 2000 Elsevier Science Ltd. All rights reserved.
引用
收藏
页码:1605 / 1615
页数:11
相关论文
共 20 条
[1]   Consistency and asymptotic normality of some subspace algorithms for systems without observed inputs [J].
Bauer, D ;
Deistler, M ;
Scherrer, W .
AUTOMATICA, 1999, 35 (07) :1243-1254
[2]  
BAUER D, IN PRESS AUTOMATICA
[3]  
BAUER D, 1995, AUTOMATICA, V31, P1865
[4]   EDGEWORTH CORRECTION BY BOOTSTRAP IN AUTOREGRESSIONS [J].
BOSE, A .
ANNALS OF STATISTICS, 1988, 16 (04) :1709-1722
[5]  
CHATTERJEE S, 1986, IEEE T SYST MAN CYB, V16, P294, DOI 10.1109/TSMC.1986.4308952
[6]   FAST SUBSPACE-BASED SYSTEM-IDENTIFICATION - AN INSTRUMENTAL VARIABLE APPROACH [J].
CHO, YM ;
KAILATH, T .
AUTOMATICA, 1995, 31 (06) :903-905
[7]  
EFRON B, 1993, INTOR BOOTSTRAP
[8]  
LARIMORE W, 1994, P 10 IFAC S SYST ID
[9]  
LOVERA M, 1998, THESIS POLITECNICO M
[10]  
OVERSCHEE P. VAN, 1996, SUBSPACE IDENTIFICAT