Numerical study of parameter identifiability for nonlinear models

Autor:	A. A. Grigor'ev, V. G. Gorskii, F. D. Klebanova, E. A. Katsman
Rok vydání:	1987
Předmět:	Nonlinear system Matrix (mathematics) symbols.namesake Basis (linear algebra) Chemistry Jacobian matrix and determinant symbols Identifiability Partial derivative Applied mathematics General Chemistry
Zdroj:	Theoretical and Experimental Chemistry. 23:181-186
ISSN:	1573-935X 0040-5760
DOI:	10.1007/bf00534579
Popis:	A method is proposed for numerical examination of identifiability for unknown parameters in nonlinear models such as occur widely in simulating reaction kinetics. The method is based on determining the basis columns in a numerical Jacobi matrix (the matrix for the partial derivatives of the model response with respect to the parameters) together with the relations of the nonbasic columns to the basic ones. One can thus determine which parameters are identified unambiguously, which are estimated in combinations, the number of such combinations, and the variances and covariances for the unambiguously identifiable parameters. Examples illustrate the method.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::b1fb2ba98968f038e90714854d43870f https://doi.org/10.1007/bf00534579 Zobrazit plný text záznamu Full text from SpringerLink