Hostname: page-component-84b7d79bbc-lrf7s Total loading time: 0 Render date: 2024-07-27T21:05:55.307Z Has data issue: false hasContentIssue false
  • English
  • Français

ESTIMATING LINEAR DYNAMICAL SYSTEMS USING SUBSPACE METHODS

Published online by Cambridge University Press:  08 February 2005

Dietmar Bauer
Dietmar Bauer
Affiliation:
TU Wien
Get access Rights & Permissions [Opens in a new window]

Abstract

This paper provides a survey on a class of so-called subspace methods whose main proponent is CCA proposed by Larimore (1983, Proceedings of the 1983 American Control Conference 2). Because they are based on regressions these methods for the estimation of ARMAX systems are attractive as a result of their conceptual simplicity and their numerical advantages as compared to traditional estimators based on criterion optimization. Under the assumption of correct specification the methods provide consistent and asymptotically normal estimates for stationary ARMAX processes where the innovations may be conditionally heteroskedastic and the exogenous inputs are strictly stationary of sufficiently short memory. For stationary autoregressive moving average (ARMA) processes with independent and identically distributed (i.i.d.) Gaussian innovations the estimates are even asymptotically efficient. For I(1) ARMA processes the estimates of both the long-run and the short-run dynamics are consistent without using the knowledge that the data are integrated in the algorithm. Additionally the algorithms provide easily accessible information on the appropriateness of the chosen model complexity. The algorithms include a number of design parameters that have to be set by the user. These include the order of the estimated system. This paper collects up-to-date knowledge on the effects of these design parameters, leading to a number of suggested automated choices to obtain a fully automated estimation procedure.

Type
Research Article
Information
Econometric Theory , Volume 21 , Issue 1 , February 2005 , pp. 181 - 211
Copyright
© 2005 Cambridge University Press

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

REFERENCES

Anderson, T.W. (1951) Estimating linear restrictions on regression coefficients for multivariate normal distributions. Annals of Mathematical Statistics 22, 327–351. Correction, Annals of Statistics 8 (1980), 1400.Google Scholar
Aoki, M. & A. Havenner (1991) State space modeling of multiple time series. Econometric Reviews 10, 1–59.Google Scholar
Baillie, R. (1996) Long memory processes and fractional integration in econometrics. Journal of Econometrics 73, 5–59.Google Scholar
Bauer, D. (1998) Some Asymptotic Theory for the Estimation of Linear Systems Using Maximum Likelihood Methods or Subspace Algorithms. Ph.D. thesis, TU Wien, Austria.
Bauer, D. (2000) Comparing the CCA Subspace Method to Pseudo Maximum Likelihood Methods in the Case of No Exogenous Inputs. Technical report, Institute of Econometrics, Operations Research and System Theory, TU Wien.
Bauer, D. (2001) Order estimation for subspace methods. Automatica 37, 1561–1573.Google Scholar
Bauer, D. (2004a) Asymptotic properties of subspace estimators. Automatica (forthcoming).Google Scholar
Bauer, D. (2004b) Estimation and Testing in Cointegrated Systems Using Subspace Methods. Technical report, Cowles Foundation for Research in Economics, Yale University.
Bauer, D. (2004c) Using Subspace Methods for Estimating ARMA Models for Multivariate Time Series with Conditionally Heteroskedastic Innovations. Technical report CFDP 1452. Cowles Foundation for Research in Economics, Yale University.
Bauer, D., M. Deistler, & W. Scherrer (1999) Consistency and asymptotic normality of some subspace algorithms for systems without observed inputs. Automatica 35, 1243–1254.Google Scholar
Bauer, D., M. Deistler, & W. Scherrer (2000) On the impact of weighting matrices in subspace algorithms. In R. Smith (ed.), Proceedings of the IFAC Conference SYSID'00. Elsevier.
Bauer, D. & M. Jansson (2000) Analysis of the asymptotic properties of the MOESP type of subspace algorithms. Automatica 36, 497–509.Google Scholar
Bauer, D. & L. Ljung (2002) Some facts about the choice of the weighting matrices in Larimore type of subspace algorithms. Automatica 38, 763–773.Google Scholar
Bauer, D. & S. de Waele (2003) A finite sample comparison of automatic model selection methods. In P. van den Hof, B. Wahlberg, & S. Weiland (eds.), Proceedings of the IFAC Symposium on System Identification, SYSID'03, Rotterdam, The Netherlands. Elsevier.
Bauer, D. & M. Wagner (2002a) Estimating cointegrated systems using subspace algorithms. Journal of Econometrics 111, 47–84.Google Scholar
Bauer, D. & M. Wagner (2002b) A Canonical Form for Unit Root Processes in the State Space Framework, Technical report, TU Wien, Austria.
Camba-Mendez, G. & G. Kapetanios (2001) Testing the rank of the Hankel covariance matrix: A statistical approach. IEEE Transactions on Automatic Control 46, 331–336.Google Scholar
Chatelin, F. (1983) Spectral Approximation of Linear Operators. Academic Press.
Chiuso, A. & G. Picci (2003) The asymptotic variance of subspace estimates. Journal of Econometrics 118, 257–291.Google Scholar
Chui, N. (1997) Subspace methods and informative experiments for system identification. Ph.D. thesis, Cambridge University.
Dahlen, A. & W. Scherrer (2004) The relation of the CCA subspace method to a balanced reduction of an autoregressive model. Journal of Econometrics 118, 293–312.Google Scholar
Deistler, M., K. Peternell, & W. Scherrer (1995) Consistency and relative efficiency of subspace methods. Automatica 31, 1865–1875.Google Scholar
Durbin, J. (1960) The fitting of time series models. Review of International Institute of Statistics 28, 233–244.Google Scholar
Hannan, E.J. & M. Deistler (1988) The Statistical Theory of Linear Systems. Wiley.
Ho, B. & R.E. Kalman (1966) Efficient construction of linear state variable models from input/output functions. Regelungstechnik 14, 545–548.Google Scholar
Jansson, M. & B. Wahlberg (1998) On consistency of subspace methods for system identification. Automatica 34, 1507–1519.Google Scholar
Kuersteiner, G. (2005) Automatic inference for infinite order vector autoregressions. Econometric Theory (this issue).Google Scholar
Larimore, W.E. (1983) System identification, reduced order filters and modeling via canonical variate analysis. In H.S. Rao & P. Dorato (eds.), Proceedings of the 1983 American Control Conference 2, pp. 445–451. IEEE.
Lewis, R. & G. Reinsel (1985) Prediction of multivariate time series by autoregressive model fitting. Journal of Multivariate Analysis 16, 393–411.Google Scholar
Peternell, K. (1995) Identification of Linear Dynamic Systems by Subspace and Realization-Based Algorithms. Ph.D. thesis, TU Wien.
Peternell, K., W. Scherrer, & M. Deistler (1996) Statistical analysis of novel subspace identification methods. Signal Processing 52, 161–177.Google Scholar
Shibata, R. (1981) An optimal autoregressive spectral estimate. Annals of Statistics 9, 300–306.Google Scholar
Sorelius, J. (1999) Subspace-Based Parameter Estimation Problems in Signal Processing. Ph.D. thesis, Uppsala University, Sweden.
Van Overschee, P. & B. DeMoor (1994) N4SID: Subspace algorithms for the identification of combined deterministic-stochastic systems. Automatica 30, 75–93.Google Scholar
Verhaegen, M. (1994) Identification of the deterministic part of mimo state space models given in innovations form from input-output data. Automatica 30, 61–74.Google Scholar
Wagner, M. (2004) A comparison of Johansen's, Bierens' and the subspace algorithm method for cointegration analysis. Oxford Bulletin of Economics and Statistics 66, 399–424.Google Scholar
Zeiger, H.P. & A.J. McEwen (1974) Approximate linear realizations of given dimension via Ho's algorithm. IEEE Transaction on Automatic Control 19, 153.Google Scholar