SeSDA: Solution path for semiparametric sparse discriminant analysis

In TULIP: A Toolbox for Linear Discriminant Analysis with Penalties

Description

Compute the solution path for semiparametric sparse discriminant analysis.

Usage

SeSDA(x,y,standardize=FALSE,lambda=NULL,alpha=1,eps=1e-7)

Arguments

x	Input matrix of predictors. x is of dimension N*p; each row is an observation vector.
y	An n-dimensional vector containing the class labels. The classes have to be labeled as 1 and 2.
standardize	A logic object indicating whether x should be standardized after transformation but before fitting classifier. Default is FALSE.
lambda	A sequence of lambda's. If lambda is missed or NULL, the function will automatically generates a sequence of lambda's to fit model.
alpha	The elasticnet mixing parameter, the same as in glmnet. Default is alpha=1 so that the lasso penalty is used.
eps	Convergence threshold for coordinate descent, the same as in glmnet. Default is 1e-7.

Value

transform	The tranformation functions.
objdsda	A DSDA object fitted on transformed data.

Author(s)

Yuqing Pan, Qing Mai, Xin Zhang

References

Mai, Q., Zou, H. and Yuan, M. (2013). A direct approach to sparse discriminant analysis in ultra-high dimensions. Biometrika, 99, 29-42.

Examples

  data(GDS1615)   ##load the prostate data
  x<-GDS1615$x
  y<-GDS1615$y
  x=x[which(y<3),]
  y=y[which(y<3)]
  obj.path <- SeSDA(x,y)

TULIP documentation built on Jan. 13, 2021, 3:14 p.m.