Advertisement

Asymptotic analysis of prediction error identification methods

  • M. H. A. Davis
  • R. B. Vinter
Part of the Monographs on Statistics and Applied Probability book series (MSAP)

Abstract

Chapter 4 provided for the most part merely a description of identification methods for dynamical systems. It is true that if we limit attention to simple moving-average models with uncorrelated disturbances and if we assume that the system is describable within the model set, then the models can be reformulated as static models to which the analysis of Section 4.3 is applicable and we can deduce certain properties of the estimates. However, the question remains open of how good are the estimates when more complicated models are considered, or when the model set does not contain a description of the system. The analysis which follows is centred on this question.

Keywords

Identification Criterion Asymptotic Analysis Infinite Order True System Closed Unit Disc 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Caines, P. E. (1976) Prediction error methods for stationary stochastic processes. IEEE Trans. Automatic Control, AC-21, 500 – 506.Google Scholar
  2. Dunsmuir, W. and Hannan, E. J. (1976) Vector linear time series models. Adv. Applied Probability, 8, 339–364. (see also corrections and extensions provided by M. Deistler, W. Dunsmuir and E. J. Hannan, ibid., 10(1978), 360 – 372 ).Google Scholar
  3. Hannan, E. J. (1973) The asymptotic theory of linear time series models. J. Applied Probability, 10, 130 – 145.CrossRefGoogle Scholar
  4. Kushner, H. J. and Clarke, D. S. (1978) Stochastic Approximation Methods for Constrained and Unconstrained Systems, Springer, New York.Google Scholar
  5. Ljung, L. (1976) On the consistency of prediction error identification methods. In System Identification: Advances and Case Studies(eds R. K. Mehra and D. G. Lainiotis) Academic Press, New York.Google Scholar
  6. Ljung, L. (1977) Analysis of recursive algorithms. IEEE Trans. Automatic Control, AC-22(4), 551 – 575.Google Scholar
  7. Ljung, L. (1978) Convergence analysis of parametric identification methods. IEEE Google Scholar
  8. Trans. Automatic Control, AC-23(5), 770–783.Google Scholar
  9. Ljung, L. and Caines, P. E. (1979) Asymptotic normality of prediction error estimators for approximate system models. Stochastics, 3(1) 29 – 46.Google Scholar
  10. Solo, V. (1979) The convergence of AML. IEEE Trans. Automatic Control, AC- 24(6), 958 – 962.Google Scholar
  11. Wald, A. (1949) Note on the consistency of the maximum likelihood estimate. Ann. Math. Statistics, 20, 595 - 601.CrossRefGoogle Scholar
  12. Walker, A. M. (1964) Asymptotic properties of least estimates of parameters of the spectrum of a stationary non-deterministic time series. J. Australian Math. Soc., 4, 363 – 388.CrossRefGoogle Scholar
  13. Whittle, P. (1961) Gaussian estimation in stationary time series. Bull. Inst. Int. Statist., 33, 1 – 26.Google Scholar

Copyright information

© M. H. A. Davis and R. B. Vinter 1985

Authors and Affiliations

  • M. H. A. Davis
    • 1
  • R. B. Vinter
    • 1
  1. 1.Department of Electrical EngineeringImperial CollegeLondonUK

Personalised recommendations