Summary
A review of the studies concerning the finite sample breakdown point (BP) of the trimmed likelihood (TL) and related estimators based on the d—fullness technique of Vandev (1993), and Vandev and Neykov (1998) is made. In particular, the BP of these estimators in the frame of the generalized linear models (GLMs) depends on the trimming proportion and the quantity N(X) introduced by Müller (1995). A faster iterative algorithm based on resampling techniques for derivation of the TLE is developed. Examples of real and artificial data in the context of grouped logistic and log-linear regression models are used to illustrate the properties of the TLE.
* The authors would like to thank the editors and the referees for their helpful comments. The research of N. Neykov was partialy funded by Deutsche Forschungsgemeinschaft while he was Visiting Fellow of the Inst. of Math. Stochastics, Univ. of Gottingen during May-June 1999. The Austrian Ministry of Science fully supported the participation of N. Neykov at the International Conference on Robust Statistics, 23-27 July 2001, Vorau, Austria. He acknowledged gratefully the financial supports.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
D.V. Atanasov. About the finite sample breakdown point of the WTL estimators. Master’s thesis, Faculty of Mathematics, Sofia University, 1998. In Bulgarian.
D.V. Atanasov and N.M. Neykov. On the finite sample breakdown point of the weighted trimmed likelihood estimators and the d-fullness of a set of continuous functions. In S. Aivazian, Y. Kharin, and H. Rieder, editors, Proceedings of the CDAM Conference, volume 1, pages 52–57. Minsk, Belarus, 10–14 September 2001.
T. Bednarski and B.R. Clarke. Trimmed likelihood estimation of location and scale of the normal distribution. Austral. J. Statist.,35:141–153, 1993.
R.J. Carroll and S. Pederson. On robustness in the logistic regression model. J. Royal Statist. Soc. B, 55:693–706, 1993.
A. Christmann. Least median of weighted squares in logistic regression with large strata. Biometrika, 81:413–417, 1994.
A. Christmann and P.J. Rousseeuw. Measuring overlap in logistic regression. Computational Statistics and Data Analysis, 28:65–75, 2001.
J.B. Copas. Binary regression models for contaminated data (with discussion). J. Royal Statist. Soc. B, 50:225–265, 1988.
P.J. Green. Iteratively reweighted least squares for maximum likelihood estimation, and some robust and resistant alternatives. J. Royal Statist. Soc. B, 46:149–192, 1984.
A. Hadi and A. Luceño. Maximum trimmed likelihood estimators: A unified approach, ex-amples and algorithms. Computational Statistics and Data Analysis, 25:251–272, 1997.
D.J. Hand, F. Daly, A.D. Lunn, K.J. Mc Conway, and E. Ostrowski. A Handbook of small data sets. Chapman and Hall, New York, 1994.
O. Hössjer. Rank-based estimates in the linear model with high breakdown point. J. Am. Statist. Assoc., 89:149–158, 1994.
M. Hubert. The breakdown value of the L1 estimator in contingency tables. Statistics and Probab. Letters, 33:419–425, 1997.
R.I. Jennrich and R.H. Moore. Maximum likelihood estimation by means of nonlinear least squares. In Proceedings of the Statistical Computing Section of the Am. Statist. Assoc., pages 57–65, 1975.
H.R. Künsch, L.A. Stefanski, and R.J. Carroll. Conditionally unbiased bounded-influence estimation in general regression models, with applications to generalized linear models. J. Am. Statist. Assoc., 84:460–466, 1989.
M. Markatou, A. Basu, and B.G. Lindsay. Weighted likelihood estimating equations: The discrete case with applications to logistic regression. J. Statist. Planning and Inf., 57: 215–232,1997.
L. Mili and C.W. Coakley. Robust estimation in structured linear regression. Ann. Statist., 15:2593–2607, 1996.
C.H. Müller. Breakdown points for designed experiments. J. Statist. Planning and Inf., 45: 413–427,1995.
C.H. Müller. Robust planning and analysis of experiments. In Lecture Notes in Statistics, volume 124. Springer, New York, 1997.
C.H. Müller and N.M. Neykov. Breakdown points of the trimmed likelihood and related estimators in generalized linear models. J. Statist. Planning and Inf. Accepted, 2002.
N.M. Neykov. Robust methods with high breakdown in the multivariate statistical analysis. PhD thesis, Faculty of Mathematics, Sofia University, 1995. In Bulgarian.
N.M. Neykov and P.N. Neytchev. A robust alternative of the maximum likelihood estimator. In Short Communications of COMPSTAT’90. Dubrovnik, pages 99–100,1990.
P.N. Neytchev, N.M. Neykov, and V.K. Todorov. User’s manual of REGRESS PC program system for fitting models to data. Technical report, National Inst. of Meteorology and Hydrology, Sofia, 1994.
R.J. O’Hara Hines and E.M. Carter. Improved added variable and partial residual plots for the detection of influential observations in generalized linear models. Appl. Statist.,42: 3–20,1993.
P.J. Rousseeuw. Least median of squares regression. J. Am. Statist. Assoc., 79:871–880, 1984.
P.J. Rousseeuw and A.M. Leroy. Robust regression and outlier detection. Wiley, New York, 1987.
P.J. Rousseeuw and K. Van Driessen. Computing LTS regression for large data sets. Technical report, University of Antwerp, 1998. Submitted.
A.J. Stromberg and D. Ruppert. Breakdown in nonlinear regression. J. Am. Statist. Assoc., 87:991–997, 1992.
D.L. Vandev. A note on breakdown point of the least median squares and least trimmed squares. Statistics and Probab. Letters, 16:117–119, 1993.
D.L. Vandev and M. Marincheva. The BP of the WLT estimators in the general elliptic family of distribution. In D.L. Vandev, editor, Proceedings of Statistical Data Analysis, pages 25–31, Varna, 1996.
D.L. Vandev and N.M. Neykov. Robust maximum likelihood in the Gaussian case. In S. Morgenthaler, E. Ronchetti, and W.A. Stahel, editors, New Directions in Data Analysis and Robustness, pages 259–264. Birkhäuser, Basel, 1993.
D.L. Vandev and N.M. Neykov. About regression estimators with high breakdown point. Statistics, 32:111–129, 1998.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Neykov, N.M., Müller, C.H. (2003). Breakdown Point and Computation of Trimmed Likelihood Estimators in Generalized Linear Models. In: Dutter, R., Filzmoser, P., Gather, U., Rousseeuw, P.J. (eds) Developments in Robust Statistics. Physica, Heidelberg. https://doi.org/10.1007/978-3-642-57338-5_24
Download citation
DOI: https://doi.org/10.1007/978-3-642-57338-5_24
Publisher Name: Physica, Heidelberg
Print ISBN: 978-3-642-63241-9
Online ISBN: 978-3-642-57338-5
eBook Packages: Springer Book Archive