New algorithms for computing the least trimmed squares regression estimator

Computational Statistics and Data Analysis

Volumn 36, Issue 4, 2001, Pages 425-439

  Agulló, José ^a  

^a UNIVERSITY OF ALICANTE   (Spain)

Author keywords

Branch and bound; Exchange algorithm; High breakdown point; Least trimmed squares; Multiple linear regression; Outliers

Indexed keywords

ALGORITHMS; LEAST SQUARES APPROXIMATIONS; REGRESSION ANALYSIS;

BRANCH-AND-BOUND METHODS; LEAST TRIMMED SQUARES (LTS) ESTIMATOR;

DECISION THEORY;

ALGORITHM; REGRESSION;

EID: 0035963359     PISSN: 01679473     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0167-9473(00)00056-6     Document Type: Article

References (19)

1. Speeding up the computation of the least quartile difference estimator
2. Simulated annealing for the detection of multiple outliers using least squares and least median of squares fitting
3. Generalized S-estimators
4. Recursive residuals
5. Algorithm AS75. Basic procedures for large, sparse or weigthed linear least squares problems
6. Hedonic prices and the demand for clean air
7. The feasible solution algorithm for least trimmed squares regression
8. Location of several outliers in multiple regression data using elemental sets
9. Improved feasible solution algorithms for high breakdown estimation
10. Algorithm AS 282. High breakdown regression and multivariate estimation
11. Exact computation of the least trimmed squares estimate in simple linear regression
12. Algorithm AS 274
13. A branch and bound algorithm for feature subset selection
14. Least median of squares regression
15. A fast algorithm for highly robust regression in data mining