Conditions for attaining the global minimum in maximum likelihood system identification

Y. Zou; W.P. Heath

doi:10.3182/20090706-3-FR-2004.00184

Conditions for attaining the global minimum in maximum likelihood system identification

Y. Zou
, W.P. Heath

University of Manchester

Allbwn ymchwil: Cyfraniad at gyfnodolyn › Erthygl › adolygiad gan gymheiriaid

Crynodeb

Maximum likelihood estimation(MLE) is a popular technique in both open and closed loop identification. However when the landscape of likelihood function has several local minima, gradient based optimization might end up with local convergence. To avoid this, various non-local-minimum conditions are derived in this paper. Here we consider different model structures, in particular Output-Error, ARMAX, and Box-Jenkins models

Iaith wreiddiol	Anadnabyddus
Tudalennau (o-i)	1110-1115
Nifer y tudalennau	6
Cyfnodolyn	IFAC Proceedings Volumes (IFAC-PapersOnline)
Cyfrol	42
Rhif cyhoeddi	10
Dyddiad ar-lein cynnar	19 Chwef 2010
Dynodwyr Gwrthrych Digidol (DOIs)	https://doi.org/10.3182/20090706-3-FR-2004.00184
Statws	Cyhoeddwyd - 21 Ebr 2016
Cyhoeddwyd yn allanol	Ie

Mynediad at Ddogfen

10.3182/20090706-3-FR-2004.00184

Dyfynnu hyn

@article{8a090e2b0d7941a6985dec96fbc4d146,

title = "Conditions for attaining the global minimum in maximum likelihood system identification",

abstract = "Maximum likelihood estimation(MLE) is a popular technique in both open and closed loop identification. However when the landscape of likelihood function has several local minima, gradient based optimization might end up with local convergence. To avoid this, various non-local-minimum conditions are derived in this paper. Here we consider different model structures, in particular Output-Error, ARMAX, and Box-Jenkins models",

author = "Y. Zou and W.P. Heath",

year = "2016",

month = apr,

day = "21",

doi = "10.3182/20090706-3-FR-2004.00184",

language = "Anadnabyddus",

volume = "42",

pages = "1110--1115",

journal = "IFAC Proceedings Volumes (IFAC-PapersOnline)",

number = "10",

}

TY - JOUR

T1 - Conditions for attaining the global minimum in maximum likelihood system identification

AU - Zou, Y.

AU - Heath, W.P.

PY - 2016/4/21

Y1 - 2016/4/21

N2 - Maximum likelihood estimation(MLE) is a popular technique in both open and closed loop identification. However when the landscape of likelihood function has several local minima, gradient based optimization might end up with local convergence. To avoid this, various non-local-minimum conditions are derived in this paper. Here we consider different model structures, in particular Output-Error, ARMAX, and Box-Jenkins models

AB - Maximum likelihood estimation(MLE) is a popular technique in both open and closed loop identification. However when the landscape of likelihood function has several local minima, gradient based optimization might end up with local convergence. To avoid this, various non-local-minimum conditions are derived in this paper. Here we consider different model structures, in particular Output-Error, ARMAX, and Box-Jenkins models

U2 - 10.3182/20090706-3-FR-2004.00184

DO - 10.3182/20090706-3-FR-2004.00184

M3 - Erthygl

VL - 42

SP - 1110

EP - 1115

JO - IFAC Proceedings Volumes (IFAC-PapersOnline)

JF - IFAC Proceedings Volumes (IFAC-PapersOnline)

IS - 10

ER -