论文信息 - Linear system identification from non-stationary cross-sectional data

Linear system identification from non-stationary cross-sectional data

The identification of time invariant linear stochastic systems from cross-sectional data on non-stationary system behavior is considered. A strong consistency and asymptotic normality result for maximum likelihood and prediction error estimates of the system parameters, system and measurement noise covariances and the initial state covariance is proven. A new identifiability property for the system model is defined and appears in the set of conditions for this result. The non-stationary stochastic realization (i.e., covariance factorization) theorem in [1] describes sufficient conditions for the identifiability property to hold. An application illustrating the use of a computer program implementing the identification method is presented.

P. Caines | P. Caines | R. Goodrich | R. Goodrich

[1] L. Ljung,et al. Extended Levinson and Chandrasekhar equations for general discrete-time linear estimation problems , 1978 .

[2] Peter E. Caines,et al. On the non-stationary covariance realization problem , 1978 .

[3] M. Appelbaum,et al. Developmental changes in mental performance. , 1973, Monographs of the Society for Research in Child Development.

[4] L. Ljung,et al. Prediction error estimators: Asymptotic normality and accuracy , 1976, 1976 IEEE Conference on Decision and Control including the 15th Symposium on Adaptive Processes.

[5] A. Wald. Note on the Consistency of the Maximum Likelihood Estimate , 1949 .

[6] E. Tse,et al. On the identifiability of parameters , 1971, CDC 1971.

[7] L. W. Sontag,et al. Mental growth and personality development : a longitudinal study , 1958 .

[8] Peter E. Caines,et al. Maximum likelihood estimation of parameters in multivariate Gaussian stochastic processes (Corresp.) , 1974, IEEE Trans. Inf. Theory.

[9] P. Caines. Prediction error identification methods for stationary stochastic processes , 1976 .

[10] T. Rothenberg. Identification in Parametric Models , 1971 .

[11] P. Caines,et al. Necessary and sufficient conditions for local second order identifiability , 1978, 1978 IEEE Conference on Decision and Control including the 17th Symposium on Adaptive Processes.