Abstract
The least trimmed squares estimator (LTS) is a well known robust estimator in terms of protecting the estimate from the outliers. Its high computational complexity is however a problem in practice. We show that the LTS estimate can be obtained by a simple algorithm with the complexity 0( N In N) for large N, where N is the number of measurements. We also show that though the LTS is robust in terms of the outliers, it is sensitive to the inliers. The concept of the inliers is introduced. Moreover, the Generalized Least Trimmed Squares estimator (GLTS) together with its solution are presented that reduces the effect of both the outliers and the inliers.
Similar content being viewed by others
References
J. Chen, G. Gu, Control Oriented System Identification: Approach, John Wiley and Sons,New York: 1999.
E. Fogel, Y. Huang, On the value of information in system identification-bounded noise case, Automatica, Vol.18, No.2, pp.229–238, 1982.
L Ljung, System Identification: Theory for the Users, Prentice-Hall, Englewood CliffS, NJ, 1987.
B. M. Ninness, G. C. Goodwin, Rapproachement between bounded-error and stochastic estimation theory, Int. J. Adap. Contr. and Signal Proc., Vol.9, No.l, pp.107–132,1995.
T. Soderstrom and P. Stoica, System Identification, Prentice-Hall, Englewood Cliffi, NJ,1989.
E. W. Bai, H. Cho, R. Tempo, Y. Ye, Optimization with few violated constraints for linear bounded error parameter estimation, IEEE Trans. on Automatic Control, Vol. 47, No. 7, pp. 1060 -1077,2002.
H. Lahanier, E. Walter, R. Gomeni. OMNE: a new robust membership set estimator for the parameter of nonlinear models, J. Pharm. Biopharm., Vol.15, No.2, pp.203 - 219,1987.
P. Rousseeuw, A. Leroy, Robust Regression and Outlier Detection, John Wiley & Sons, New York, 1987.
N. Neykov, P. Neytchev, Least median of squares, least trimmed squares and S-estimations by means of BMDP3R and BMDPAR, Compu. Statistics Quarterly, Vol. 16, No. 4, pp. 281 - 293,1991.
A. Stromberg, Computing the exact least median of squares estimate and stability diagnostics in multiple linear regression, SIAM J. Sci. Comput., Vol.14, No.6, pp. 1289- 1299,1993.
P. Tichavsky, Algorithms for and geometrical characterization of solutions in the LMS and the LTS linear regression, Compu. Statistics Quarterly, Vol.6, No.2, pp. 139 - 151,1991.
L. Davis, The asymptotics of S-estimators in the linear regression model, Anna, of Statistics, Vol.18, No.4, pp. 1651 - 1675,1990.
D. Ruppert, R. Carroll, Trimmed least squares estimation in the linear model, /. American Statis. Assoc., Vol.75, No.372, pp.828 -838,1980.
R. Tempo, E. Bai, F. Dabbene, Probabilistic robustness analysis; explicit bound for the minimum number of samples, System & Control Letters, Vol.30, No.5, pp.237 - 242,1997.
Author information
Authors and Affiliations
Additional information
This work was supported in part by NSF ECS — 9710297 and ECS — 0098181.
Rights and permissions
About this article
Cite this article
Bai, E. Outliers, inliers and the generalized least trimmed squares estimator in system identification. J. Control Theory Appl. 1, 17–27 (2003). https://doi.org/10.1007/s11768-003-0004-4
Issue Date:
DOI: https://doi.org/10.1007/s11768-003-0004-4