mutoss (version 0.1-9)

paired.marginal: Marginal paired two sample test

Description

The robust version uses the Wilcoxon signed rank test, otherwise a paired t-test will be performed.

Usage

paired.marginal(data, model, robust, alternative, psi0, equalvar)

Arguments

data

the data set

model

the result of a call to paired.model(classlabel)

robust

a logical variable indicating whether a paired t-test or a Wilcoxon signed rank test should be used.

alternative

a character string specifying the alternative hypothesis, must be one of two.sided, greater or less

psi0

a numeric that defines the hypothesized null value

equalvar

a logical variable indicating whether to treat the two variances as being equal

Value

• pValues
  A numeric vector containing the unadjusted pValues

Details

A vector of classlabels needs to be provided to distinguish the two paired groups. The arrangement of group indices does not matter, as long as the columns are arranged in the same corresponding order between groups. For example, if group 1 is code as 0 and group 2 is coded as 1, for 3 pairs of data, it does not matter if the classlabel is coded as (0,0,0,1,1,1) or (1,1,1,0,0,0) or (0,1,0,1,0,1) or (1,0,1,0,1,0), the paired differences between groups will be calculated as group2 - group1.

References

Wilcoxon, F. (1945). Individual Comparisons by Ranking Methods. Biometrics Bulletin 1:80-83.