ldarob: Robust Linear Discriminant Analysis

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

Description

This model utilizes robust estimation in order to improve the performance of the standard linear discriminant. This is a heavy modification of MASS's lda function. Due to the substantial modifications the returned object is of its own "ldarob" class and has an accompanying predict method.

Usage

ldarob(
  formula,
  data,
  prior = proportions,
  method = c("rocke", "shr"),
  tol = 1e-10
)

Arguments

formula	a model formula
data	a data frame
prior	a vector of prior probabilities for each class
method	the scale estimation method. one of "rocke" (cov.rocke) or "shr" (cov.shr)
tol	tolerance for small-variance variables. If the variance is smaller than tol, the variable is declared constant. Defaults to 1e-04.

Value

An ldarob object

Examples

ldarob(Species ~.,iris)

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