Robust model selection criteria for robust S and LT S estimators

Robust model selection criteria for robust S and LT S estimators

Outliers and multi-collinearity often have large influence in themodel/variable selection process in linear regression analysis. To investigate this combined problem of multi-collinearity and outliers, westudied and compared Liu-type S (liuS-estimators) and Liu-type LeastTrimmed Squares (liuLTS) estimators as robust model selection criteria. Therefore, the main goal of this study is to select subsets of independent variables which explain dependent variables in the presence ofmulti-collinearity, outliers and possible departures from the normalityassumption of the error distribution in regression analysis using thesemodels.

___

  • Arslan, O. and Billor, N., Robust Liu estimator for regression based on an M-estimator, Journal of Applied Stats., Vol. 27, No.1, 39-47, 2000.
  • Çetin, M., Variable Selection Criteria in Robust Regression, PhD. Thesis, University of Hacettepe, 2000.
  • Çetin, M., Robust model selection criteria for robust Liu estimator, European Journal of Operation Research" , Vol.199, issue 1,21-24, 2009.
  • Erar, A., Variable Selection in Linear Regression Models in case of multicollinearity, Un- published Ph.D. Thesis, Ankara,Turkey, 1982.
  • Gunst, R. F., Mason, R.L., Advantages of examining multicollinearities in regression anal- ysis, Biometrics, 33, 249-260, 1977.
  • Hampel, F. R., Some Aspects of Model Choice in Robust Statistics, Proceedings of the th Session of the ISI, Madrid, Book 2, 767-771, 1983.
  • Hampel, F. R., Ronchetti, E. M., Rousseeuw, P. J., and Stahel, W. A., Robust Statistics: The Approach Based on Influence Functions, New York, John Wiley, 1986.
  • Hurvich, C., F. and Tsai, C. L., Model selection for least absolute deviations regression in small samples, Statistics and Probab. Letters, 9, 259-265, 1990.
  • Liu, K., A new class of baised estimate in linear regression, Comm. in Stats. A, 22,393-402, Mallows, C. L., Some comment on Cp, Technometrics, Vol.15, No: 4, 661- 675, 1973.
  • Ronchetti, E., Robust model selection in regression, Statistics and Prob. Letters., 3, 21-23, Ronchetti, E. and Staudte, R., A Robust version of Mallows's Cp, JASA,Vol. 89, 550-559, Ronchetti, E., Field, C., and Blanhard, W., Robust linear model selection by cross- validation, JASA, Vol.92, No.439, Theory and methods, 1017-1023, 1997.
  • Rousseeuw, P. J., Least Median of Squares Regression, Journal of the American Statistical Association, 79, 871-880, 1984.
  • Rousseeuw, P. J. and Yohai, V. J., Robust Regression by Means of S-estimators, in Robust and Nonlinear Time Series Analysis, ed. By W.H. Franke, and D. Martin, Springer-Verlag, New York, pp.256-272, 1984.
  • Sommer, S. and Huggins, R. M., Variable selection using the Wald test and a robust Cp, Appl. Statist., 45, 15-29, 1996.