Effect of leverage on the finite sample efficiencies of high breakdown estimators
The presence of leverage points in regression can have a dramatic impact on the finite sample efficiencies (versus least squares under Gaussian errors) of high breakdown estimators. In general, as the xvalue of a leverage point becomes more extreme, the finite sample efficiencies decrease and larger sample sizes are required for asymptotic results to apply. However, the least median of squares (LMS) and least trimmed squares (LTS) estimators exhibit higher finite sample than asymptotic efficiencies. On the other hand, the finite sample efficiency of the Schweppetype onestep generalized M estimator (S1S), starting from LTS, which minimizes the sum of the h smallest squared residuals, converges from below to its asymptotic value as the sample size increases. We suggest that the finite sample variance itself be replaced by a trimmed sample variance which accounts for the long tail behavior encountered in finite samples. An improvement on the S1S estimator, which consists of never downweighting any point with one of the h smallest squared LTS residuals, performs quite well with respect to either measure of efficiency, while retaining a high breakdown point.
Year of publication: 
1994


Authors:  Coakley, Clint W. ; Mili, Lamine ; Cheniae, Michael G. 
Published in: 
Statistics & Probability Letters.  Elsevier, ISSN 01677152.  Vol. 19.1994, 5, p. 399408

Publisher: 
Elsevier 
Keywords:  Asymptotic efficiency finite sample efficiency breakdown point exact fit point least median of squares least trimmed squares onestep estimator 
Saved in favorites
Similar items by person

Exact fit points under simple regression with replication
Coakley, Clint W., (1993)

A Bounded Influence, High Breakdown, Efficient Regression Estimator
Coakley, Clint W., (1993)

Economic market design and planning for electric power systems
Momoh, James, (2010)
 More ...