The DKL-optimality criterion has been recently proposed for the dual problem of model discrimination and parameter estimation, for the case of two rival models. A sequential version of the DKL-optimality criterion is herein proposed in order to discriminate and efficiently estimate more than two nested non-linear models. Our sequential method is inspired by the procedure of Biswas and Chaudhuri (2002), which is however useful only :in the set up of nested linear models.
An adaptive sequential design for model discrimination and parameter estimation in non-linear nested models / C. Tommasi, C. May (AIP CONFERENCE PROCEEDINGS). - In: International Conference of Numerical Analysis and Applied Mathematics / [a cura di] G. Psihoyios, C. Tsitouras. - [s.l] : AIP, 2010. - ISBN 9780735408340. - pp. 1917-1920 (( convegno ICNAAM tenutosi a Rhodes nel 2010 [10.1063/1.3498294].
An adaptive sequential design for model discrimination and parameter estimation in non-linear nested models
C. Tommasi;
2010
Abstract
The DKL-optimality criterion has been recently proposed for the dual problem of model discrimination and parameter estimation, for the case of two rival models. A sequential version of the DKL-optimality criterion is herein proposed in order to discriminate and efficiently estimate more than two nested non-linear models. Our sequential method is inspired by the procedure of Biswas and Chaudhuri (2002), which is however useful only :in the set up of nested linear models.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.