roblm.lqd: Least Quartile Difference Regression GS-Estimator

In abnormally-distributed/cvreg: Cross Validation and Robust Estimation Utilities

Description

Note this is best used as an initial S-estimate rather than as an estimator itself. While more efficient than LTS and LMS, this still has less than 80

Usage

roblm.lqd(formula, data)

Arguments

formula	model formula
data	data frame
k	the trimming percentage. the default amount of trimming is 0.30.
h	the number of cases to retain. If h is specfied, use h smallest residuals, and ignore k.

Value

a roblm object

References

Croux, C., Rousseeuw, P. & Hössjer, O. (1994) Generalized S-Estimators. Journal of the American Statistical Association, 89:428, 1271-1281, DOI: 10.1080/01621459.1994.10476867

abnormally-distributed/cvreg documentation built on May 3, 2020, 3:45 p.m.