lrperm {logregperm}R Documentation

Deviance for a Logistic Regression with Permuted Variables

Description

lrperm permutes the residuals from a linear regression of the independent variable of interest on the remaining independent variables, and then computes the deviance from a logistic regression of these residuals and the other independent variables on the dependent variable. This function is intended to be used only by prr.test

Usage

lrperm(y, xx, resid, size)

Arguments

y the dependent binary variable; a vector.
xx independent variables about which inference is not to be made in a matrix in which rows correspond to observations and each column to a variable. xx must include a vector of 1s for the intercept.
resid a vector of residuals from a linear regression of the variable of interest on the remaining variables (in xx).
size number of observations used.

Value

lrperm returns the deviance as described above

Author(s)

Douglas M. Potter

References

Potter D.M. (2005) A permutation test for inference in logistic regression with small- and moderate-sized datasets. Statistics in Medicine, 24:693-708.

Examples

nobs<-40
x1<-rnorm(nobs)
x2<-rnorm(nobs)
xx<-cbind(x1,x2)
xint<-rep(1,nobs)

x0<-rnorm(nobs)+x1+x2
y<-x0+x1+x2+2*rnorm(nobs)
y<-ifelse(y>0,1,0)

resid<-lm(x0~x1+x2)$residuals

lrperm(y,cbind(xint,xx),resid,nobs)

## The function is currently defined as
function(y,xx,resid,size){
sel<-sample(1:size,rep=F)
x<-cbind(resid[sel],xx)
return(glm.fit(x,y,family=binomial())$deviance)
  }

[Package logregperm version 1.0 Index]