demo.WilcoxonMannWhitney: Generate sample data that is appropriate for the classical...

Description Usage Arguments Value

Description

Generate sample data that is appropriate for the classical distributionfree tests

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
demo.WilcoxonMannWhitney(n = 100, groupmeans = c(0, 1),
  force.balanced = FALSE)

demo.KruskalWallis(n = 100, groupmeans = c(0, 1, 2),
  force.balanced = FALSE)

demo.MackSkillings(blockeffects = rnorm(4), groupmeans = c(0, 1, 2),
  replications = 10, force.balanced = FALSE)

demo.JonckheereTerpstra(n = 100, groupmeans = c(0, 1, 2, 4),
  force.balanced = FALSE)

demo.MackWolfe(n = 100, groupmeans = c(0, 1, 5, 2),
  force.balanced = FALSE)

demo.BrownHettmansperger(n = 100, groupmeans = c(0, 1, 2),
  force.balanced = FALSE)

Arguments

n

Number of observations

groupmeans

Means of normal outcome distribution for each group (note this implies the number of groups)

force.balanced

If TRUE, the design is forced to be balanced (an error occurs if this is impossible)

blockeffects

Effect of each block (note this implies the number of blocks)

replications

Number of replications per group/block combination

Value

a list holding the following items:

dta

A data.frame holding the generated data. The outcome is column "y", the predictor is "x" and the blocking variable (if relevant) is "b"

n

As passed in.

groupmeans

As passed in.

For demo.MackSkillings these additional items are present:

blocks

Number of blocks.

blockeffects

(Random) effect of each specific block.

treatments

Number of groups in the predicting variable.

tb1

data.frame holding all combinations of groups and blocks.

tbmult

Similar as tb1, but now holding the repetitions and also randomized.


pimold documentation built on May 2, 2019, 5:50 p.m.