Analysis of a Covariance Matching Method for Discrete-Time Errors-in-Variables Identification

A covariance matching method for discrete-time errors-in-variables identification is analyzed. The method consists of fitting covariance and cross-covariance functions, parameterized by the unknown parameters, to covariance and cross-covariance functions estimated from the noise-corrupted data. An approximative expression for the covariance matrix of the estimated parameter vector is given.