Maximum Likelihood Recursive Extended Least Squares Estimation for Multivariate Equation-Error Moving Average Systems

Abstract

In this paper, the parameter identification of the multivariate equation-error moving average system is studied. The system is decomposed into several subsystems and a maximum likelihood recursive extended least squares identification algorithm is presented for estimating the parameter vector in each subsystem. The numerical simulation results indicate that the maximum likelihood recursive extended least squares identification algorithm is effective and get more accurate parameter estimates.