Roorda B., Heij C. Global Total Least Squares Modelling of Multivariable Time Series. IEEE Transcations on Automatic Control, Vol.40, no.1, pp.50-63, 1995. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
On the formal equivalence between static and dynamic.. - Lemmerling, Dologlou.. (1998) (Correct)
....which especially in the dynamic case is often overlooked. Key words: modeling, ARX, dynamic Errors In Variables 1 Introduction In this correspondence we clarify the equivalences between the Auto Regressive models with eXogenous input (ARX) 6] and the dynamic Errors In Variables (EIV) models [10,3,7,8]. Therefore, we first focus on their static counterparts, namely the Least Squares (LS) model [4] and the Total Least Squares (TLS) model [9,4] We show how a TLS problem can be written in the formalism of 1 This work was supported by the Belgian Programme on Interuniversity Poles of Attraction ....
Roorda B., Heij C. Global Total Least Squares Modelling of Multivariable Time Series. IEEE Transcations on Automatic Control, Vol.40, no.1, pp.50-63, 1995.
Estimation of Factor Models by Realization-Based and.. - Scherrer Self-citation (Heij) (Correct)
....between inputs and outputs and we do not require that all the variables are restricted by the system. We consider two identification methods within this setting. The first method uses an explicit identification criterion corresponding to (nonlinear) least squares approximation of the data (see [6, 10, 12]) The second method is based on the statistical concept of principal components (see [1] and uses approximate realization algorithms in the spirit of subspace identification (see [8, 15] Our analysis extends earlier results in [13] a b c d e ARMA Y Y Y Y polynomial IO PEI N N Y Y ....
....The methods BLS, IRM and PFM identify behaviours B with corresponding errors minfEkw(t) Gamma w(t)k 2 ; w 2 Bg. This is equal to trace Z Gamma (I Gamma PB )S(I Gamma PB )d (6) where S is the process spectrum and PB is the orthogonal projection operator onto the behaviour B (see [12] for computational details) The means and standard deviations in Table 2 refer to this criterion. The results (and other simulation experiments) suggest that BLS gives a good estimate in general. However, for complex models BLS is quite time consuming and the faster algorithms IRM and PFM do a ....
B. Roorda and C. Heij. Global total least squares modelling of multivariable time series. IEEE Transactions on Automatic Control 40:50-63, 1995.
Consistency of System Identification by Global Total Least.. - Heij, Scherrer (1996) (2 citations) Self-citation (Heij) (Correct)
No context found.
Roorda, B. and C. Heij (1995). Global total least squares modelling of multivariable time series. IEEE Transactions on Automatic Control 40, pp. 50-63.
Identification of System Behaviours by Approximation of Time.. - Scherrer, Heij (1997) Self-citation (Heij) (Correct)
....Earlier roots of these ideas are in [1, 2] for stochastic systems. Another approach is more equation oriented as in [20] see for example [6, 17] and this related to prediction error methods [9] A third approach, based on least squares approximation of behaviours, is developed in [11] see also [8, 13]. In this paper we consider the behavioural least squares method (BLS) also called global total least squares. In a sense, this method adheres most strictly to the behavioural principle of focusing on the external system characteristics. Subspace methods and realization based methods for data ....
B. Roorda and C. Heij, Global total least squares modelling of multivariable time series, IEEE Transactions on Automatic Control 40, 1995, pp. 50-63.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC