A High Breakdown and Efficient GM-Estimator in Linear Models

  • Song, Moon-Sup (Department of Statistics, Seoul National University) ;
  • Park, Changsoon (Department of Applied Statistics. Chung-Ang University) ;
  • Nam, Ho-Soo (Department of Industrial Engineering, Dongseo University)
  • 발행 : 1996.12.01

초록

In this paper we propose an efficient scoring type one-step GM-estimator, which has a bounded influence function and a high break-down point. The main point of the estimator is in the weighting scheme of the GM-estimator. The weight function we used depends on both leverage points and residuals So we construct an estimator which does not downweight good leverage points Unider some regularity conditions, we compute the finite-sample breakdown point and prove asymptotic normality Some simulation results are also presented.

키워드

참고문헌

  1. The Theory of Linear Models and Multivariate Analysis Arnold, S. F.
  2. Journal aof the American Statistical Association v.70 One-Step Huber Estimates in the Linear Model Bickel, P. J.
  3. Journal of the American Statistical Association v.88 A Bounded Influence, High Breakdown, Effcient Regression Estimator Coakley, C. W.;Herrmansperger, T. P.
  4. Algorithms, Routines, and S Functions for Robust Statistics Marazzi, A.
  5. Zeitschrift fut Wahrscheinlichkeitstheorire und vermeandte Gebiete v.58 Asymptotic Behavior of General M-estimates for Regression and Scale with Random Carriers Maronna, R. A.;Yohai, V> J.
  6. Journal of the American Statistical Association v.79 Least Median of Squares Regression Rousseeuw, P. J.
  7. Approximation Theorems of Mathmetical Statistics Serfling, R. J.
  8. Journal of the American Statistical Association v.87 On One-Strp GM Estimates and Stability of Inferences in Linear Regression Simpson, D. G.;Ruppert, D.;Carroll, R. J.
  9. S-PLUS for Windows User's Manual Statististifcal Sciences