文章基本信息

标题：Affinely Parametrized State-space Models: Ways to Maximize the Likelihood Function
本地全文：下载
作者：Adrian Wills ; Chengpu Yu ; Lennart Ljung 等
期刊名称：IFAC PapersOnLine
印刷版ISSN：2405-8963
出版年度：2018
卷号：51
期号：15
页码：718-723
DOI：10.1016/j.ifacol.2018.09.170
语种：English
出版社：Elsevier
摘要：AbstractUsing Maximum Likelihood (or Prediction Error) methods to identify linear state space model is a prime technique. The likelihood function is a nonconvex function and care must be exercised in the numerical maximization. Here the focus will be on affine parameterizations which allow some special techniques and algorithms. Three approaches to formulate and perform the maximization are described in this contribution: (1) The standard and well known Gauss-Newton iterative search, (2) a scheme based on the EM (expectation-maximization) technique, which becomes especially simple in the affine parameterization case, and (3) a new approach based on lifting the problem to a higher dimension in the parameter space and introducing rank constraints.
关键词：KeywordsParameterized state-space modelmaximum-likelihood estimationexpectation-maximization algorithmdifference-of-convex optimization