Theory and methods: Small sample performance of extreme regression quantiles

Simple item page
dc.contributor.authorDe Jongh P.J.
dc.contributor.authorDe Wet T.
dc.date.accessioned2011-05-15T16:02:32Z
dc.date.available2011-05-15T16:02:32Z
dc.date.issued2003
dc.description.abstractIn this paper we studied the finite sample behaviour of a range of estimators for extreme regression quantiles. We studied the behaviour of the estimators over a wide range of heavy tailed error distributions and design matrices containing influential design points. Our main conclusion is that restricted bounded influence estimators should be the estimators of choice. In particular the preferred estimator is a restricted bounded influence Koenker-Bassett based on a common slope estimator of a 20% trimmed mean. This estimator has good performance in well-behaved cases and protects one against the adverse effects of heavy tails and high leverage design points.
dc.description.versionArticle
dc.identifier.citationSouth African Statistical Journal
dc.identifier.citation37
dc.identifier.citation2
dc.identifier.issn0038271X
dc.identifier.urihttp://hdl.handle.net/10019.1/12513
dc.titleTheory and methods: Small sample performance of extreme regression quantiles
dc.typeArticle
Files
Collections
Stellenbosch University - Scopus Publications