treeperm: Permutation Tests Using Tree Algorithms

Description

This function will perform a K samnple permutation test of data locations. Users may choose either asymptotic test or exact test.

Usage

"treeperm"(formula, frame = list(), type, size, ...)
"treeperm"(x,data,factor, type, size, ...)
"print"(x,...)
"summary"(object,...)

Arguments

Generic argument, used only in print.treeperm to represent the treeperm class.

formula

The formula to perform permutation test. The form should be value~label, where value is the set of data and label classifies value into groups.

frame

The dataframe storing vectors used in formula.

data

The data set to be permuted

factor

The index set that classifies parameter data into groups.

type

Which type of permutation the user want to use, either 'exact' or 'approximate'.

size

If the type is 'approximate', user must specify the size of permutations they want.

object

The treeperm class to be summarised.

...

Reserved for future.

Value

result: A permutation object returned by a permutation function, either class Permutation or class KPermutation depending the levels of argument factor.
call: User's calling script to this function at highest level.
data: The input data set.
factor: The index set that classifies data into groups.

Details

This function calls KPermutation or Permutation depending on the levels of factor.

Examples

Run this code

value<-c(0,190,0,0,10,0,0,0,0,0,0,110,0,0,52,0,8,0,50,0,0,137,965,110)
label<-as.factor(c(1,1,1,2,2,2,1,1,1,2,2,2,1,1,1,2,2,2,1,1,1,2,2,2))
data<-data.frame(label,value)
exact<-treeperm(value~label,frame=data,type="exact")
summary(exact)
approximate<-treeperm(value~label,frame=data,type="approximate",size=999)
summary(approximate)