computeTestPartialRegression {rotRPackage}R Documentation

Compute the Partial Regression Test on 2 data sets.

Description

This ROT function, called from a Test C++ object, is given 2 samples, a selection and optionnaly a test level. It then computes the result of a regression test against the null hypothesis that the selected linear regression coefficients are significant. It returns the test result for each selected X variable, the statistics and the p-values.

Usage

computeTestPartialRegression(outSample, inSample, selection, testLevel = 0.95)

Arguments

outSample The 'out' sample. (n vector)
inSample The 'in' sample (m-by-n matrix).
selection The partial variable selection (vector of column indexes)
testLevel the test level. (scalar in [0:1])

Value

A list is returned, containing :

testResult The result. 1 means H0 is not rejected. (vector)
threshold The threshold applied to the p-value when deciding the outcome of the test.
pValues The test pValues. (vector)

Author(s)

Pierre-Matthieu Pair, Softia for EDF.

Examples

outSample=iris[1:50,1]
inSample=as.matrix(iris[1:50,2:4])
computeTestPartialRegression(outSample, inSample, c(1,2))

[Package rotRPackage version 1.4.3 Index]