R Graphical Manual

Browse All

Last data update: 2014.03.03

R: Calculate statistics and p-values

Pvalue

R Documentation

Calculate statistics and p-values

Description

Calculate F-statistics, t-statistics and corresponding p-values given multiple regression models under the null and alternative hypotheses.

Usage

  Pvalue(dat, mod, mod0)

Arguments

`dat`	An N genes by n arrays matrix of expression data
`mod`	An n by (s+1) design matrix under the full model (alternative), the first column is the primary predictor, s>=0 and the rest of the columns are additional covariates
`mod0`	An n by s design matrix under the null hypothesis, s>=0, should be the same as the 2:(s+1) columns of mod. If s=0, please set mod0 = `NULL`

Value

`p`	An N by 1 vector of p-values one for each row of data.
`tstat`	An N by 1 vector of t-statistics for primary parameters.
`fstat`	An N by 1 vector of F-statistics for primary parameters.
`coef`	An N by (s+1) matrix of coefficients with respect to design matrix mod under the full model.

Author(s)

Yunting Sun yunting.sun@gmail.com, Nancy R.Zhang nzhang@stanford.edu, Art B.Owen owen@stanford.edu

Examples

  ## Not run: 
      data(simdat)
      n = ncol(simdat$data)
      mod = cbind(simdat$g, rep(1,n))
      mod0 = cbind(rep(1,n))
      result = Pvalue(dimdat$data,mod = mod, mod0 = mod0)
  
## End(Not run)