uvExactNull: Exact univariate null distribution

In MaStatLab/MultiFit: Multiscale Fisher's Independence Test for Multivariate Dependence

Description

In a univariate case, simulate an exact null distribution for the global test statistics.

Usage

uvExactNull(uv.null.sim, test.method, correct = TRUE, col0.tot, row0.tot,
  grand.tot, top.max.ps = 4L, verbose = FALSE)

Arguments

uv.null.sim	Positive integer, the number of simulated values to be computed in a univariate case when an exact or approximate null distribution is simulated.
test.method	String, "Fisher" and "norm.approx" are applicable here.
correct	Logical, relates to Fisher's exact test, if TRUE the exact null is simulated for the mid-p corrected p-values.
col0.tot	Numerical vector, containing the column-0 totals of all 2x2 contingency tables for which simulated values are asked for. Has to be the same length as row0.tot and grand.tot.
row0.tot	Numerical vector, containing the row-0 totals of all 2x2 contingency tables for which simulated values are asked for. Has to be the same length as col0.tot and grand.tot.
grand.tot	Numerical vector, containing the totals of all 2x2 contingency tables for which simulated values are asked for. Has to be the same length as col0.tot and row0.tot.
top.max.ps	Positive integer, report the mean of the top top.max.ps order statistics of the negative logarithm of all p-values.
verbose	Logical.

Value

List of two numerical vectors for each of the global test statistics. Each such vector is of length uv.null.sim.

MaStatLab/MultiFit documentation built on Jan. 11, 2020, 5:11 p.m.