pamr.cv {pamr}R Documentation

A function to cross-validate the nearest shrunken centroid classifier

Description

A function to cross-validate the nearest shrunken centroid classifier produced by pamr.train

Usage

pamr.cv(fit, data,  nfold = NULL, folds = NULL,...)

Arguments

fit The result of a call to pamr.train
data A list with at least two components: x- an expression genes in the rows, samples in the columns), and y- a vector of the class labels for each sample. Same form as data object used by pamr.train.
nfold Number of cross-validation folds. Default is the smallest class size
folds A list with nfold components, each component a vector of indices of the samples in that fold. By default a (random) balanced cross-validation is used
...

{Any additional arguments that are to be passed to pamr.train}

Details

pamr.cv carries out cross-validation for a nearest shrunken centroid classifier.

Value

A list with components

threshold
errors The number of cross-validation errors for each threshold value
loglik The cross-validated multinomial log-likelihood value for each threshold value
size A vector of the number of genes that survived the thresholding, for each threshold value tried.
yhat A matrix of size n by nthreshold, containing the cross-validated class predictions for each threshold value, in each column
prob A matrix of size n by nthreshold, containing the cross-validated class probabilities for each threshold value, in each column
folds The cross-validation folds used
cv.objects Train objects (output of pamr.train), from each of the CV folds
call The calling sequence used

Author(s)

Trevor Hastie,Robert Tibshirani, Balasubramanian Narasimhan, and Gilbert Chu

References

Examples

set.seed(120)
x <- matrix(rnorm(1000*20),ncol=20)
y <- sample(c(1:4),size=20,replace=TRUE)

mydata <- list(x=x,y=factor(y), geneid=as.character(1:nrow(x)), genenames=paste("g",as.character(1:nrow(x)),sep=""),)

mytrain <-   pamr.train(mydata)
mycv <- pamr.cv(mytrain,mydata)

[Package pamr version 1.28.0 Index]