Recursive maximum likelihood identification of a non-linear output-affine model

From MaRDI portal





This is a readable paper the subject of which is the estimation of parameters in a non-linear output-affine model by using a recursive maximum likelihood (RML) algorithm. The structure of the model is assumed to be known. The basic principles of the RML can be readily extended from the linear case to the non-linear case. Thus, both the derivation and the general convergence analysis of the RML algorithm introduced in the paper are straightforward. However, some ``details in the analysis, such as the invertibility of the noise model, are more complicated in the non-linear case and they are only briefly analysed in this paper. The paper includes some numerical applications to both simulated and real data, showing the typical performance that can be achieved by the derived RML algorithm.











This page was built for publication: Recursive maximum likelihood identification of a non-linear output-affine model

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3807999)