ltsbase-package: Ridge and Liu Estimates based on LTS Method
In ltsbase: Ridge and Liu Estimates based on LTS (Least Trimmed Squares) Method

This is a package that gives the estimates of Ridge and Liu parameters based on LTS method in multiple linear regression analysis. It can be also used to compare the biasing parameters obtained from Ridge regression, Ridge based on LTS method, Liu, and Liu based on LTS method. It measures the performance of the models according to MSE and extract the biasing parameter at minimum MSE. Additionally, it is possible to compare the MSE values of the four regression models on a plot.

Package:	ltsbase
Type:	Package
Version:	1.0.1
Date:	2013-08-02
License:	GPL-3

ltsbase package has a main function called ltsbase and also two other useful functions called ltsbaseDefault and ltsbaseSummary. The function ltsbase computes the minimum MSE values for Ordinary Least Squares, Ridge, Ridge based on LTS, LTS, Liu, Liu based on LTS method. The second function ltsbaseDefault is used to get the fitted values and residuals of the corresponding model. The last function ltsbaseSummary is used to get the regression coefficients and the biasing parameter for the best MSE among four regression models.

Betul Kan Kilinc <bkan@anadolu.edu.tr>; Ozlem Alpu <oalpu@ogu.edu.tr>. We are grateful to Berna Yazici for various suggestions and contributions.

ltsbase documentation built on May 2, 2019, 8:31 a.m.