Xmc.sevsample: Generalized Wald-type Statistics: Several Sample RAD...

In HMP: Hypothesis Testing and Power Calculations for Comparing Metagenomic Samples from HMP

Description

This function computes the Generalized Wald-type test statistic (Wilson and Koehler, 1984) and corresponding p-value to assess whether the sample RAD probability-means from multiple populations are the same or different. The statistics assumes that a common RAD probability-mean vector for comparison under the null hypothesis is known.

Usage

Xmc.sevsample(group.data, pi0)

Arguments

group.data	A list where each element is a matrix of taxonomic counts(columns) for each sample(rows).
pi0	The RAD-probability mean vector.

Details

Note: The matrices in group.data must contain the same taxa, in the same order.

Value

A list containing the Generalized Wald-type statistics and p-value.

References

Wilson, J. R., and Koehler, K. J. (1984). Testing of equality of vectors of proportions for several cluster samples. Proceedings of Joint Statistical Association Meetings. Survey Research Methods.

Examples

	data(saliva) 
	data(throat)
	data(tonsils)
	
	### Get pi from the dirichlet-multinomial parameters
	pi0 <- dirmult(saliva)$pi
	
	### Combine the data sets into a single list
	group.data <- list(throat, tonsils)
	
	xmc <- Xmc.sevsample(group.data, pi0)
	xmc

Example output

Loading required package: dirmult

Attaching package: 'HMP'

The following object is masked from 'package:dirmult':

    weirMoM

Iteration 1: Log-likelihood value: -1426219.55743915
Iteration 2: Log-likelihood value: -1426159.69392069
Iteration 3: Log-likelihood value: -1426137.57581644
Iteration 4: Log-likelihood value: -1426134.36208055
Iteration 5: Log-likelihood value: -1426134.28420196
Iteration 6: Log-likelihood value: -1426134.28414898
$`Xmc statistics`
[1] 295.529

$`p value`
[1] 0

HMP documentation built on Aug. 31, 2019, 5:05 p.m.