pamr.fdr: A function to estimate false discovery rates for the nearest...

In pamr: Pam: Prediction Analysis for Microarrays

A function to estimate false discovery rates for the nearest shrunken centroid classifier

Description

A function to estimate false discovery rates for the nearest shrunken centroid classifier

Usage

pamr.fdr(
  trained.obj,
  data,
  nperms = 100,
  xl.mode = c("regular", "firsttime", "onetime", "lasttime"),
  xl.time = NULL,
  xl.prevfit = NULL
)

Arguments

trained.obj	The result of a call to pamr.train
data	Data object; same as the one passed to pamr.train
nperms	Number of permutations for estimation of FDRs. Default is 100
xl.mode	Used by Excel interface
xl.time	Used by Excel interface
xl.prevfit	Used by Excel interface

Details

pamr.fdr estimates false discovery rates for a nearest shrunken centroid classifier

Value

A list with components:

results	Matrix of estimates FDRs for various various threshold values. Reported are both the median and 90th percentile of the FDR over permutations
pi0	The estimated proportion of genes that are null, i.e. not significantly different

Author(s)

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

Examples

suppressWarnings(RNGversion("3.5.0"))
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)
myfdr <- pamr.fdr(mytrain, mydata)

pamr documentation built on July 1, 2024, 9:06 a.m.