mtest

The goal of mtest is to study contingency tables. It is related to Fisher's exact test and to Barnard's exact test. The main difference with Barnard's test is that the nuisance parameter is not estimated, but integrated over all possible values.

Installation

You can use devtools to install from this repository:

install_github('vqf/mtest', build_vignettes = T)

You will be able to install the released version of mtest shortly from CRAN with:

install.packages("mtest")

Example

This is a basic example which shows you how to solve a common problem:

library(mtest)
m.test(list(c(10, 7), c(11, 6)))
tailed.m.test(list(c(10, 7), c(11, 6)))