Computing the Least Quartile Difference Estimator in the Plane
A common problem in linear regression is that largely aberrant values can strongly influence the results. The least quartile difference (LQD) regression estimator is highly robust, since it can resist up to almost 50% largely deviant data values without becoming extremely biased. Additionally, it shows good behavior on Gaussian data – in contrast to many other robust regression methods. However, the LQD is not widely used yet due to the high computational effort needed when using common algorithms, e.g. the subset algorithm of Rousseeuw and Leroy. For computing the LQD estimator for n data points in the plane, we propose a randomized algorithm with expected running time O(n2 log2 n) and an approximation algorithm with a running time of roughly O(n2 log n). It can be expected that the practical relevance of the LQD estimator will strongly increase thereby.
Year of publication: |
2005
|
---|---|
Authors: | Bernholt, Thorsten ; Nunkesser, Robin ; Schettlinger, Karen |
Publisher: |
Dortmund : Universität Dortmund, Sonderforschungsbereich 475 - Komplexitätsreduktion in Multivariaten Datenstrukturen |
Saved in:
freely available
Series: | Technical Report ; 2005,51 |
---|---|
Type of publication: | Book / Working Paper |
Type of publication (narrower categories): | Working Paper |
Language: | English |
Other identifiers: | 510723926 [GVK] hdl:10419/22644 [Handle] RePEc:zbw:sfb475:200551 [RePEc] |
Source: |
Persistent link: https://www.econbiz.de/10010296715
Saved in favorites
Similar items by person
-
Computing the Least Quartile Difference Estimator in the Plane
Bernholt, Thorsten, (2005)
-
Computing the least quartile difference estimator in the plane
Bernholt, Thorsten, (2007)
-
Computing the least quartile difference estimator in the plane
Bernholt, Thorsten, (2005)
- More ...