perm-internal: Internal Functions for perm package: Not to be called by User

Description

These functions are where the algorithms are done. There is much room for improvement in the speed of the exact functions.

Usage

ksample.exact.mc(scores, group, nmc = 10^4 - 1, seed = 1234321, 
  digits = 12, p.conf.level = 0.99, setSEED = TRUE)
ksample.pclt(scores, group)
trend.exact.mc(scores, group, alternative = "two.sided", nmc = 10^3 - 1, 
  seed = 1234321, digits = 12, p.conf.level = 0.99, setSEED = TRUE)
trend.pclt(scores, group)
twosample.exact.ce(scores, group, cm = NULL, digits = 12)
twosample.exact.mc(scores, group, alternative = "two.sided", nmc = 10^4 - 1, 
  seed = 1234321, digits = 12, p.conf.level = 0.99, setSEED = TRUE)
twosample.pclt(scores, group)
twosample.exact.network(scores, group, digits = 12)
getcnt(nodehk, cnt.edge, edgesize)

Value

The function getcnt returns a vector of indeces representing arcs for set of nodes

Arguments

scores: vector of response scores
group: covariate vector
alternative: one of 'less', 'greater', 'two.sided' or 'two.sidedAbs'
nmc: number of Monte Carlo replications
seed: random number seed
digits: digits for rounding of test statistic, equal to that many digits are called tied
p.conf.level: confidence level for p-value, used with mc methods
setSEED: logical, set to FALSE when performing simulations on mc methods
cm: for speed you can input the matrix created from chooseMatrix (see chooseMatrix). If NULL it is created.
nodehk: nodes for which indeces of arcs are needed
cnt.edge: vector of first index for each node
edgesize: vector of number of arcs for each node

Details

Network algorithm is very basic, only works for two group tests. The function getcnt (called by twosample.exact.network) gets a vector of indeces representing arcs for set of nodes.