pamr (version 1.25)

pamr.cv: 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

...

Value

• A list with components
• threshold
• {A vector of the thresholds tried in the shrinkage}
• 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

Details

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

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)