DocumentCode :
2700042
Title :
A new robust estimation approach: An extended threshold M-estimator procedure
Author :
Corbier, C. ; Carmona, J.-C. ; Alvarado, V.A.
Author_Institution :
Lab. des Sci. de l´´Inf. et des Syst., ENSAM, Aix-en-Provence, France
fYear :
2011
fDate :
26-28 Oct. 2011
Firstpage :
1
Lastpage :
6
Abstract :
In order to tackle more efficiently the parameters estimation of an Output Error (OE) models contaminated by outliers, we propose to extend the range of the scaling factor of a parameterized robust estimation criterion (PREC) in the Huber´s M-estimates context based on a mixed norm. Moreover, since the gradient and the Hessian of the PREC present a nonlinear structure in the OE models, we propose a new method to establish an L-Finite Taylor´s Expansion of these expressions in order to provide the asymptotic covariance matrix of the robust estimator. We present the results of a Monte Carlo study and we compare some robust methods with respect to our procedure.
Keywords :
Hessian matrices; Monte Carlo methods; covariance matrices; estimation theory; parameter estimation; time series; Huber M-estimates; L-finite Taylor expansion; Monte Carlo study; asymptotic covariance matrix; extended threshold M-estimator procedure; output error model; parameter estimation; parameterized robust estimation criterion; robust estimation approach; scaling factor; time series; Computational modeling; Covariance matrix; Estimation; Robustness; Taylor series; Time series analysis; Vectors; L-Finite Taylor´s Expansion; OE models; asymptotic covariance matrix; outliers; robust estimation; scaling factor;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electrical Engineering Computing Science and Automatic Control (CCE), 2011 8th International Conference on
Conference_Location :
Merida City
Print_ISBN :
978-1-4577-1011-7
Type :
conf
DOI :
10.1109/ICEEE.2011.6106691
Filename :
6106691
Link To Document :
بازگشت