Mtest: Perform M-Test on RVG Frequency Table
In rvgtest: Tools for Analyzing Non-Uniform Pseudo-Random Variate Generators

Description Usage Arguments Details Value Author(s) References See Also Examples

Perform an adjusted residual test (“M-Test”) on the RVG frequency table.

1	rvgt.Mtest(ftable)

ftable

object of class "rvgt.ftable".

rvgt.Mtest performs an adjusted residual test (“M-Test”) on the bin counts of the RVG frequency tables created by means of rvgt.ftable. The null hypothesis is that the bin counts are distributed according to the expected probabilities (i.e., proportional to the lengths of the histogram cells).

If the ftable contains the bin counts for more than one repetitions then the test is repeated with column sums of increasing number of rows which corresponds to increasing sampling sizes. This allows for getting an idea of the power of the test.

An object of class "rvgt.htest" which is a list with components:

`type`	character string containing type of test; equals `"M-test"`.
`n`	sample size for every row in `ftable`.
`rep`	number of rows in `ftable`.
`breaks`	number of break points in `ftable` (i.e., number of bins + 1).
`pval`	vector of p-values.

Sougata Chaudhuri sgtchaudhuri@gmail.com, Josef Leydold josef.leydold@wu.ac.at

C. Fuchs and R. Kenett. A test for Detecting Outlying Cells in the Multinomial Distribution and Two-Way Contingency Tables. Journal of American Statistical Association, Vol 75, June 1980, 395–398.

See plot.rvgt.htest for the syntax of the plotting method.

## Create a frequency table for normal distribution and
## conduct an M-test on this data.
## Use a sample of size of 5 times 10^5 random variates.
table <- rvgt.ftable(n=1e5,rep=5, rdist=rnorm,qdist=qnorm, mean=1,sd=2)

## Perform test
result <- rvgt.Mtest(table)

## Plot result
plot(result)