galts

Genetic Algorithms and C-Steps Based LTS (Least Trimmed Squares) Estimation

Last update: 20 August 2023

Copyright license: GNU General Public License, version 3.0, GNU General Public License, version 2.0

Software version identifier: 1.3.1, 1.0, 1.1, 1.2, 1.3, 1.3.2

Source code repository: https://github.com/cran/galts

Description

Includes the ga.lts() function that estimates LTS (Least Trimmed Squares) parameters using genetic algorithms and C-steps. ga.lts() constructs a genetic algorithm to form a basic subset and iterates C-steps as defined in Rousseeuw and van-Driessen (2006) to calculate the cost value of the LTS criterion. OLS (Ordinary Least Squares) regression is known to be sensitive to outliers. A single outlying observation can change the values of estimated parameters. LTS is a resistant estimator even the number of outliers is up to half of the data. This package is for estimating the LTS parameters with lower bias and variance in a reasonable time. Version >=1.3 includes the function medmad for fast outlier detection in linear regression.