The fast-tau estimator for regression

Subscribe to email list

Please select the email list(s) to which you wish to subscribe.
CAPTCHA
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Image CAPTCHA

Enter the characters shown in the image.

User menu

You are here

  1. Department of Statistics
  2. The fast-tau estimator for regression

The fast-tau estimator for regression

TitleThe fast-tau estimator for regression
Publication TypeJournal Article
Year of Publication2008
AuthorsSalibian-Barrera, M, Willems, G, Zamar, R
JournalJOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS
Volume17
Pagination659-682
Date PublishedSEP
Type of ArticleArticle
ISSN1061-8600
Keywordsrandom resampling, Robust regression, simulated annealing, tabu search
AbstractYohai and Zamar's tau-estimators of regression have excellent statistical properties but are nevertheless rarely used in practice because of a lack of available software and the general impression that tau-estimators are difficult to approximate. We will show, however, that the computational difficulties of approximating tau-estimators are similar in nature to those of the more popular S-estimators. The main goal of this article is to compare an approximating algorithm for tau-estimators based on random resampling with some alternative heuristic search algorithms. We show that the former is not only simpler, but that when enhanced by local improvement steps it generally outperforms the considered heuristic search algorithms, even when these heuristic algorithms also incorporate local improvement steps. Additionally, we show that the random resampling algorithm for approximating tau-estimators has favorable statistical properties compared to the analogous and widely used algorithms for S- and least trimmed squares estimators.
DOI10.1198/106186008X343785