penalizedLDA (version 1.1)

penalizedLDA-package: Penalized linear discriminant analysis using lasso and fused lasso penalties.

Description

This package performs penalized linear discriminant analysis, intended for the high-dimensional setting in which the number of features p exceeds the number of observations n. Fisher's discriminant problem is modified in two ways: 1. A diagonal estimate of the within-class class covariance is used. 2. Lasso or fused lasso penalties are applied to the discriminant vectors in order to encourage sparsity, or sparsity and smoothness.

Details

Package:

penalizedLDA

Type:

Package

Version:

1.1

Date:

2015-07-09

License:

GPL (>=2.0)

LazyLoad:

yes

The main functions are PenalizedLDA, which performs penalized linear discriminant analysis, and PenalizedLDA.cv, which performs cross-validation in order to select the optimal tuning parameters for penalized LDA.

References

D Witten and R Tibshirani (2011) Penalized classification using Fisher's linear discriminant. To appear in Journal of the Royal Statistical Society, Series B.

Examples

set.seed(1)
n <- 20
p <- 100
x <- matrix(rnorm(n*p), ncol=p)
y <- c(rep(1,5),rep(2,5),rep(3,10))
x[y==1,1:10] <- x[y==1,1:10] + 2
x[y==2,11:20] <- x[y==2,11:20] - 2
out <- PenalizedLDA(x,y,lambda=.14,K=2)
print(out)