TESTS FOR LINEARITY AND BILINEARITY OF DYNAMIC SYSTEMS.

P. Várlaki, G. Terdik, V. A. Lototsky

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

In the preliminary data analysis of the identification procedure some statistical methods are necessary to decide first on the linear or nonlinear character of the systems. This decision /and the nonlinear structure estimation/ is possible only by statistical evaluation of the observed input/output data of the dynamic systems using measures of nonlinearity and tests for linearity. This paper derives new test methods for checking the linear and bi-linear character of dynamic systems in the case of the Gaussian stationary /ARMA/ input processes. The paper deals with the relationship between the above test methods and estimations for measures of nonlinearity. Using these tests and measures one can distinguish among nonlinear model structures represented by Volterra and Zadeh functional series of different orders.

Original languageEnglish
Title of host publicationIFAC Proceedings Series
PublisherPergamon Press
Pages407-432
Number of pages26
Edition7
ISBN (Print)0080325424
Publication statusPublished - 1985

Fingerprint

Dynamical systems
Model structures
Statistical methods

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Várlaki, P., Terdik, G., & Lototsky, V. A. (1985). TESTS FOR LINEARITY AND BILINEARITY OF DYNAMIC SYSTEMS. In IFAC Proceedings Series (7 ed., pp. 407-432). Pergamon Press.

TESTS FOR LINEARITY AND BILINEARITY OF DYNAMIC SYSTEMS. / Várlaki, P.; Terdik, G.; Lototsky, V. A.

IFAC Proceedings Series. 7. ed. Pergamon Press, 1985. p. 407-432.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Várlaki, P, Terdik, G & Lototsky, VA 1985, TESTS FOR LINEARITY AND BILINEARITY OF DYNAMIC SYSTEMS. in IFAC Proceedings Series. 7 edn, Pergamon Press, pp. 407-432.
Várlaki P, Terdik G, Lototsky VA. TESTS FOR LINEARITY AND BILINEARITY OF DYNAMIC SYSTEMS. In IFAC Proceedings Series. 7 ed. Pergamon Press. 1985. p. 407-432
Várlaki, P. ; Terdik, G. ; Lototsky, V. A. / TESTS FOR LINEARITY AND BILINEARITY OF DYNAMIC SYSTEMS. IFAC Proceedings Series. 7. ed. Pergamon Press, 1985. pp. 407-432
@inproceedings{27b1dde46a6c459a9c6a6460890d7205,
title = "TESTS FOR LINEARITY AND BILINEARITY OF DYNAMIC SYSTEMS.",
abstract = "In the preliminary data analysis of the identification procedure some statistical methods are necessary to decide first on the linear or nonlinear character of the systems. This decision /and the nonlinear structure estimation/ is possible only by statistical evaluation of the observed input/output data of the dynamic systems using measures of nonlinearity and tests for linearity. This paper derives new test methods for checking the linear and bi-linear character of dynamic systems in the case of the Gaussian stationary /ARMA/ input processes. The paper deals with the relationship between the above test methods and estimations for measures of nonlinearity. Using these tests and measures one can distinguish among nonlinear model structures represented by Volterra and Zadeh functional series of different orders.",
author = "P. V{\'a}rlaki and G. Terdik and Lototsky, {V. A.}",
year = "1985",
language = "English",
isbn = "0080325424",
pages = "407--432",
booktitle = "IFAC Proceedings Series",
publisher = "Pergamon Press",
edition = "7",

}

TY - GEN

T1 - TESTS FOR LINEARITY AND BILINEARITY OF DYNAMIC SYSTEMS.

AU - Várlaki, P.

AU - Terdik, G.

AU - Lototsky, V. A.

PY - 1985

Y1 - 1985

N2 - In the preliminary data analysis of the identification procedure some statistical methods are necessary to decide first on the linear or nonlinear character of the systems. This decision /and the nonlinear structure estimation/ is possible only by statistical evaluation of the observed input/output data of the dynamic systems using measures of nonlinearity and tests for linearity. This paper derives new test methods for checking the linear and bi-linear character of dynamic systems in the case of the Gaussian stationary /ARMA/ input processes. The paper deals with the relationship between the above test methods and estimations for measures of nonlinearity. Using these tests and measures one can distinguish among nonlinear model structures represented by Volterra and Zadeh functional series of different orders.

AB - In the preliminary data analysis of the identification procedure some statistical methods are necessary to decide first on the linear or nonlinear character of the systems. This decision /and the nonlinear structure estimation/ is possible only by statistical evaluation of the observed input/output data of the dynamic systems using measures of nonlinearity and tests for linearity. This paper derives new test methods for checking the linear and bi-linear character of dynamic systems in the case of the Gaussian stationary /ARMA/ input processes. The paper deals with the relationship between the above test methods and estimations for measures of nonlinearity. Using these tests and measures one can distinguish among nonlinear model structures represented by Volterra and Zadeh functional series of different orders.

UR - http://www.scopus.com/inward/record.url?scp=0022309737&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0022309737&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0022309737

SN - 0080325424

SP - 407

EP - 432

BT - IFAC Proceedings Series

PB - Pergamon Press

ER -