Model-structure selection by cross-validation

From MaRDI portal





Let \(S\) denote the system that generates the data. The data are realizations of stationary ergodic processes, \(M(\theta)\) denotes a mathematical model of \(S\), and \(\theta\) a finite-dimensional vector of unknown parameters. Two criteria for choosing between different models are proposed. These criteria asymptotically reduce to some well-known structure selection criteria. Also, a very important purpose of the paper is to given an extensive survey on other criteria.



Cites work


Cited in
(28)








This page was built for publication: Model-structure selection by cross-validation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3744063)