Maximum likelihood estimation of linear SISO models subject to missing output data and missing input data
Authors:
- 2 May 2014
- journal article
- research article
- Published by Taylor & Francis in International Journal of Control
- p. 1-11
Abstract
In this paper we describe an approach to maximum likelihood estimation of linear single input single output (SISO) models when both input and output data are missing. The criterion minimised in the algorithms is the Euclidean norm of the prediction error vector scaled by a particular function of the covariance matrix of the observed output data. We also provide insight into when simpler and in general sub-optimal schemes are indeed optimal. The algorithm has been prototyped in MATLAB, and we report numerical results that support the theory.Keywords
This publication has 22 references indexed in Scilit:
- Frequency domain system identification with missing dataIEEE Transactions on Automatic Control, 2000
- The exact likelihood of an autoregressive-moving average model with incomplete dataBiometrika, 1997
- Sequential algorithms for parameter estimation based on the Kullback-Leibler information measureIEEE Transactions on Acoustics, Speech, and Signal Processing, 1990
- Optimal ARMA parameter estimation based on the sample covariances for data with missing observationsIEEE Transactions on Information Theory, 1989
- Exact likelihood of vector autoregressive-moving average process with missing or aggregated dataBiometrika, 1983
- Estimation of Time Series Models in the Presence of Missing DataJournal of the American Statistical Association, 1981
- Maximum Likelihood Fitting of ARMA Models to Time Series With Missing ObservationsTechnometrics, 1980
- Parameter estimation for auto-regressive systems with missing observationsInternational Journal of Systems Science, 1980
- A variable projection method for solving separable nonlinear least squares problemsBIT Numerical Mathematics, 1975
- Missing Data in an Autoregressive ModelInternational Economic Review, 1974