ARMA identification
autoregressive moving average (ARMA) processesHankel matrix of moment estimatesspectral identificationsuccessive 1-step predictors
Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Inference from stochastic processes and spectral analysis (62M15) Inference from stochastic processes and prediction (62M20) Estimation and detection in stochastic control theory (93E10) Identification in stochastic control theory (93E12)
In view of recent results on the asymptotic behavior of the prediction error covariance for a state variable system, see \textit{R. S. Bucy} [Math. Syst. Theory 16, No.4, 307-317 (1983; Zbl 0538.93059)], an identification scheme for autoregressive moving average (ARMA) processes is proposed. The coefficients of the d-step predictor determine asymptotically the system moments \(U_ 0,...,U_{d-1}\). These moments are also nonlinear functions of the coefficients of the successive 1-step predictors. Here, we etimate the state variable parameters by the following scheme. First, we use the Burg technique, see \textit{J. P. Burg} [Maximum entropy spectral analysis, Ph. D. thes., Stanford Univ. 1975], to find the estimates of the coefficients of the successive 1-step predictors. Second, we compute the moments by substitution of the estimates provided by the Burg technique for the coefficients in the nonlinear functions relating the moments with the 1-step predictor coefficients. Finally, the Hankel matrix of moment estimates is used to determine the coefficients of the characteristic polynomial of the state transition matrix, see \textit{H. Padé} [Ann. Sci. Norm. Supér., III. Sér. 9, Suppl., 3-93 (1892)] and \textit{B. L.Ho} and \textit{R. E. Kalman} [Regelungstechnik 14, 545-548 (1966; Zbl 0145.127)]. A number of examples for the state variable systems corresponding to ARMA (2,1) processes are given which show the efficiency of this technique when the zeros and poles are separated. Some of these examples are also studied with an alternative technique, see \textit{M. I. Ribeiro} and \textit{J. M. F. Moura} [Tech. Rep. No.LIDS-P-1521, Massachusetts Inst. Technol. 1985], which exploits the linear dependence between successive 1-step predictors and the coefficients of the transfer function numerator and denominator polynomials. In this paper, the problems of order determination are not considered; we assumed the order of the underlying system. We remark that the Burg algorithm is a robust statistical procedure. With the notable exception of \textit{H. Akaike} [Ann. Inst. Stat. Math. 26, 363-387 (1974; Zbl 0335.62058)] who uses canonical correlation methods, most identification procedures in control are based on a deterministic analysis and consequently are quite sensitive to errors. In general, spectral identification based on the windowing of data lacks the resolving power of the Burg technique, which is a super resolution method.
- scientific article; zbMATH DE number 3233996 (Why is no real title available?)
- scientific article; zbMATH DE number 3395169 (Why is no real title available?)
- Identification and filtering
- Machine independent monte carlo evaluation of the performance of dynamic stochastic systems
- Markovian representation of stochastic processes and its application to the analysis of autoregressive moving average processes
- Mathematical Description of Linear Dynamical Systems
- The Synthesis of Linear Dynamical Systems from Prescribed Weighting Patterns
- Prediction and filtration of a partially observed vector ARMA system of first order
- Identification and filtering
- On the identification of AR systems excited by periodic signals of unknown phase
- scientific article; zbMATH DE number 1532436 (Why is no real title available?)
- AN APPLICATION OF THE SCHUR‐COHN ALGORITHM TO TIME SERIES ANALYSIS
- Identification of an unstable ARMA equation
- Uniqueness of estimated k-step prediction models of ARMA processes
This page was built for publication: ARMA identification
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1089301)
