Consistency and relative efficiency of subspace methods

被引：74

作者：

Deistler, M ^{[1
]}

Peternell, K ^{[1
]}

Scherrer, W ^{[1
]}

机构：

[1] VIENNA TECH UNIV,INST OKONOMETRIE OPERAT RES & SYST THEORIE,A-1040 VIENNA,AUSTRIA

来源：

AUTOMATICA | 1995年 / 31卷 / 12期

关键词：

linear dynamic systems; state-space representations; identification algorithms; subspace methods; Hannan-Rissanen method; maximum-likelihood estimators; statistical analysis; consistency; relative efficiency;

D O I：

10.1016/0005-1098(95)00089-6

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

We give a consistency proof for two subspace methods. We then show the asymptotic equivalence of a special subspace method and the initial estimate proposed by Hannan and Rissanen. Finally, a simulation study comparing two subspace methods and the maximum-likelihood method is performed.

引用

页码：1865 / 1875

页数：11

共 15 条

[1] STOCHASTIC THEORY OF MINIMAL REALIZATION [J].

AKAIKE, H .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1974, AC19 (06) :667-674

[2] MARKOVIAN REPRESENTATION OF STOCHASTIC-PROCESSES BY CANONICAL VARIABLES [J].

AKAIKE, H .

SIAM JOURNAL ON CONTROL, 1975, 13 (01) :162-173

[3]

Akaike H, 1976, SYSTEM IDENTIFICATIO, P27, DOI [10.1016/s0076-5392(08)60869-3, DOI 10.1016/S0076-5392(08)60869-3, 10.1016/S0076-5392(08)60869-3]

[4]

ASTROM KJ, 1982, 6TH P IFAC S ID SYST, P122

[5] STRUCTURE AND ORDER ESTIMATION OF MULTIVARIABLE STOCHASTIC-PROCESSES [J].

FUCHS, JJ .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1990, 35 (12) :1338-1341

[6]

Golub G. H., 1989, MATRIX COMPUTATIONS

[7]

HANNAN EJ, 1982, BIOMETRIKA, V69, P81, DOI 10.1093/biomet/69.1.81

[8] MULTIVARIATE LINEAR TIME-SERIES MODELS [J].

HANNAN, EJ ;

KAVALIERIS, L .

ADVANCES IN APPLIED PROBABILITY, 1984, 16 (03) :492-561

[9]

HANNAN EJ, 1988, STATISTICAL THEORY L

[10]

Larimore W. E., 1983, Proceedings of the 1983 American Control Conference, P445

← 1 2 →