This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.
Abstract Details
Activity Number:
|
357
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 3, 2010 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Statistical Learning and Data Mining
|
Abstract - #307805 |
Title:
|
A Deterministic Algorithm for the LTS
|
Author(s):
|
Tim Verdonck*+ and Mia Hubert and Peter Rousseeuw
|
Companies:
|
University of Antwerp and Katholieke Universiteit Leuven and Katholieke Universiteit Leuven
|
Address:
|
Middelheimlaan 1 (Middelheimcampus M.G.320b), Antwerp, 2020, Belgium
|
Keywords:
|
regression ;
robustness ;
outliers ;
affine equivariance
|
Abstract:
|
The least trimmed squares (LTS) regression estimate is the least squares fit to the h (out of n) observations with smallest least squares objective function. Besides being highly robust, the LTS estimates are regression, scale, and affine equivariant. Computing the exact LTS is very hard, so in practice one resorts to approximate algorithms. Most often the FASTLTS algorithm is used. This algorithm starts by drawing many initial fits on random subsets, followed by so-called concentration steps. The FASTLTS algorithm is affine equivariant, but not permutation invariant. In this article we present a deterministic algorithm, denoted as DetLTS, which does not use random subsets. We illustrate DetLTS on real and simulated data sets.
|
The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.
Back to the full JSM 2010 program
|
2010 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.