f.pvalue: A function for quickly calculating f statistic p-values for...

In steveneschrich/msva: Surrogate Variable Analysis

Description

This function does simple linear algebra to calculate f-statistics for each row of a data matrix comparing the nested models defined by the design matrices for the alternative (mod) and and null (mod0) cases. The columns of mod0 must be a subset of the columns of mod.

Usage

f.pvalue(dat, mod, mod0)

Arguments

dat	The transformed data matrix with the variables in rows and samples in columns
mod	The model matrix being used to fit the data
mod0	The null model being compared when fitting the data

Value

p A vector of F-statistic p-values one for each row of dat.

Examples

library(bladderbatch)
data(bladderdata)
dat <- bladderEset[1:50,]

pheno = pData(dat)
edata = exprs(dat)
mod = model.matrix(~as.factor(cancer), data=pheno)
mod0 = model.matrix(~1,data=pheno)

pValues = f.pvalue(edata,mod,mod0)
qValues = p.adjust(pValues,method="BH")

steveneschrich/msva documentation built on Dec. 23, 2021, 5:33 a.m.