Title :
Guaranteed linear and nonlinear parameter estimation from bounded-error data: A survey
Author :
Walter, Eric ; Piet-Lahanier, Hèlène
Author_Institution :
Lab. des Signaux et Syst., CNRS-ESE, Gif-sur-Yvette, France
Abstract :
Classical statistical estimation has some drawbacks that can be overcome by bounded-error estimation, a deterministic alternative. Methods are presented that make it possible to obtain a set guaranteed to contain all parameter vectors consistent with the data, model structure and assumed error bounds. The techniques to be used depend on whether or not the error is affine in the parameters
Keywords :
error analysis; parameter estimation; affine error deterministic methods; bounded-error data; error bounds; linear parameter estimation; model structure; nonlinear parameter estimation; parameter vectors; Bayesian methods; Biology computing; Computer aided manufacturing; Concrete; Error correction; Maximum likelihood estimation; Measurement errors; Parameter estimation; Predictive models; Probability density function;
Conference_Titel :
Circuits and Systems, 1993., ISCAS '93, 1993 IEEE International Symposium on
Conference_Location :
Chicago, IL
Print_ISBN :
0-7803-1281-3
DOI :
10.1109/ISCAS.1993.393837
