Abstract
This chapter presents and discusses various aspects of what theory predicts in terms of accuracy of instrumental variable estimates. A general derivation of the covariance matrix of the parameter estimates is presented. This matrix is influenced by a number of user choices in the identification method, and it is further discussed how these user choices can be made in order to make the covariance matrix as small as possible in a well-defined sense. The chapter includes also a comparison with the prediction error method, and a discussion of in what situations an optimal instrumental variable method can be statistically efficient.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Box, G.E.P., Jenkins, G.W.: Time Series Analysis, Forecasting and Control, 2nd edn. Holden-Day, San Francisco (1976)
Friedlander, B.: The overdetermined recursive instrumental variable method. IEEE Trans. Autom. Control AC-29, 353–356 (1984)
Gilson, M., Van den Hof, P.: Instrumental variable methods for closed-loop identification. Automatica 41(2), 241–249 (2005)
Ljung, L.: System Identification—Theory for the User, 2nd edn. Prentice Hall, Upper Saddle River (1999)
Ljung, L., Söderström, T.: Theory and Practice of Recursive Identification. MIT Press, Cambridge (1983)
Mayne, D.Q.: A method for estimating discrete time transfer functions. In: Advances in Computer Control, Second UKAC Control Convention, Bristol, UK (1967)
Peterka, V., Halousková, Q.: Tally estimate of åström model for stochastic systems. In: Proc. 2nd IFAC Symposium on Identification and System Parameter Estimation, Prague, Chechoslovakia (1970)
Reiersøl, O.: Confluence analysis by means of lag moments and other methods of confluence analysis. Econometrica 9, 1–24 (1941)
Söderström, T.: Errors-in-variables methods in system identification. Automatica 43(6), 939–958 (2007). Survey paper
Söderström, T., Hong, M.: Identification of dynamic errors-in-variables systems with periodic data. In: Proc. 16th IFAC World Congress, Prague, Czech Republic, July 4–8 (2005)
Söderström, T., Stoica, P.: Instrumental Variable Methods for System Identification. Springer, Berlin (1983)
Söderström, T., Stoica, P.: System Identification. Prentice Hall International, Hemel Hempstead (1989)
Söderström, T., Stoica, P., Trulsson, E.: Instrumental variable methods for closed loop systems. In: 10th IFAC World Congress, Munich, Germany (1987)
Stoica, P., Friedlander, B., Söderström, T.: Instrumental variable methods for ARMA models. In: Leondes, C.T. (ed.) Control and Dynamic Systems—Advances in Theory and Applications. System Identification and Adaptive Control, vol. 25, pp. 79–150. Academic Press, New York (1987)
Stoica, P., Söderström, T.: Optimal instrumental variable estimation and approximate implementation. IEEE Trans. Autom. Control AC-28, 757–772 (1983)
Stoica, P., Söderström, T., Friedlander, B.: Optimal instrumental variable estimates of the AR parameters of an ARMA process. IEEE Trans. Autom. Control AC-30, 1066–1074 (1985)
Söderström, T., Stoica, P., Friedlander, B.: An indirect prediction error method. Automatica 27, 183–188 (1991)
Thil, S., Hong, M., Söderström, T., Gilson, M., Garnier, H.: Statistical analysis of a third-order cumulants based algorithm for discrete errors-in-variables identification. In: IFAC 17th World Congress Seoul, Korea, July 6–11 (2008)
Van Huffel, S., Vandewalle, J.: Comparison of total least squares and instrumental variable methods for parameter estimation of transfer function models. Int. J. Control 50, 1039–1056 (1989)
Wong, K.Y., Polak, E.: Identification of linear discrete time systems using the instrumental variable approach. IEEE Trans. Autom. Control 12, 707–718 (1967)
Young, P.C.: An instrumental variable method for real-time identification of a noisy process. Automatica 6, 271–287 (1970)
Young, P.C., Jakeman, A.J.: Refined instrumental variable methods of time series analysis: Part III extensions. Int. J. Control 31, 741–764 (1980)
Young, P.C.: Parameter estimation for continuous-time models—a survey. Automatica 17, 23–29 (1981)
Young, P.C.: Recursive Estimation and Time-Series Analysis. Springer, Berlin (1984)
Young, P.C., Jakeman, A.J.: Refined instrumental variable methods of recursive time-series analysis. Part I: Single input, single output systems. Int. J. Control 29, 1–30 (1979)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Additional information
Dedicated to Professor Peter Young on the occasion of his 70th anniversary, with thanks for many years of discussions on the instrumental variable method.
Appendix: Proofs and Derivations
Appendix: Proofs and Derivations
1.1 A.1 Proof of Lemma 1.1
Using the assumption on joint Gaussian distribution we can apply the general rule for product of Gaussian variables
Using the result (1.82) in (1.34) leads to
Recall that the covariance function r v (τ) decays exponentially with τ. Therefore we can write
for some |α|<1. Using this result, we get
Now use the conventions
The assumption (1.36) now implies that
as at least one of the factors is zero. Therefore
which is (1.37).
1.2 A.2 Proof of Lemma 1.2
Using the definitions, one find that an arbitrary element of the matrix S is given by
To proceed we need to evaluate the expectation of products of the white noise sequence. Set m e =E{e 4(t)}. As e(t) has zero mean, the expected value of a product of four factors of the noise is nonzero if either the time arguments are pairwise equal, or all are equal. This principle gives
Inserting this into (1.89) leads to
Comparing the calculations in the proof of Lemma 1.3.1, we find that the first term in (1.91) is precisely
Further, the second term turns out to be
which vanishes due to the assumption (1.44).
The last term can be written as
However, we know that
This implies that
and the lemma is proven.
1.3 A.3 Proof of Lemma 1.3
We first write from (1.47)
The inequality (1.54) can equivalently be written as
which can be rewritten as
This in turn follows from the theory of partitioned matrices, cf. Lemma A.3 of [12], as
1.4 A.4 Proof of Lemma 1.4
Using the theory of partitioned matrices, see for example Lemma A.2 in [12] and (1.56)
Equality in (1.101) applies if and only if
The condition (1.59) is equivalent to
for some matrix α. As S is nonsingular, this is in turn equivalent to α=S −1 R, and
which is (1.102).
1.5 A.5 Answer to Exercise 1.3
Use the notation
Then
1.6 A.6 Proof of Lemma 1.5
Using the definition (1.50) of K(z) introduce the notations
Then it holds
Using (1.48) leads to
The stated inequality (1.62) then reads
Now, (1.113) is equivalent to
which follows from the theory of partitioned matrices, cf. Lemma A.4 in [12], as
Further, we see that with the specific choice
it holds that
from which the equality in (1.62) follows.
Rights and permissions
Copyright information
© 2012 Springer-Verlag London Limited
About this chapter
Cite this chapter
Söderström, T. (2012). How Accurate Can Instrumental Variable Models Become?. In: Wang, L., Garnier, H. (eds) System Identification, Environmental Modelling, and Control System Design. Springer, London. https://doi.org/10.1007/978-0-85729-974-1_1
Download citation
DOI: https://doi.org/10.1007/978-0-85729-974-1_1
Publisher Name: Springer, London
Print ISBN: 978-0-85729-973-4
Online ISBN: 978-0-85729-974-1
eBook Packages: EngineeringEngineering (R0)