Xoc.sevsample {HMP} | R Documentation |
Likelihood-Ratio-Test Statistics: Several Sample Overdispersion Test Comparison
Description
This routine provides the value of the likelihood-ratio-test statistic and the corresponding p-value to assess whether the overdispersion observed in multiple groups of microbiome samples are equal.
Usage
Xoc.sevsample(group.data, epsilon = 10^(-4))
Arguments
group.data |
A list where each element is a matrix of taxonomic counts(columns) for each sample(rows). (See Notes 1 and 2 in details) |
epsilon |
Convergence tolerance. To terminate, the difference between two succeeding log-likelihoods must be smaller than epsilon. Default value is 10^(-4). |
Details
To assess whether the over dispersion parameter vectors \theta_{\mathrm{m}}
observed in J
groups of microbiome samples are equal to each other, the following hypothesis
\mathrm{H}_{\mathrm{o}}: \theta_{\mathrm{1}} = \cdots =\theta_{\mathrm{m}} =\cdots=\theta_{\mathrm{J}}=\theta_{\mathrm{o}}
versus \mathrm{H}_{\mathrm{a}}: \theta_{\mathrm{m}} \ne \theta_{\mathrm{o}}, m=1, \ldots, J
can be tested. In particular, the likelihood-ratio test statistic is used (Tvedebrink, 2010), which is given by,
x_{\mathrm{oc}}=-2 \log\left\{\frac{L\left(\theta_{\mathrm{o}}; \mathbf{X}_{\mathrm{1}},\ldots, \mathbf{X}_{\mathrm{J}} \right)}{L\left(\theta_{\mathrm{1}},\ldots, \theta_{\mathrm{J}}; \mathbf{X}_{\mathrm{1}},\ldots, \mathbf{X}_{\mathrm{J}} \right)}\right\} .
The asymptotic null distribution of x_{\mathrm{oc}}
follows a Chi-square with degrees of freedom equal to (J-1) (Wilks, 1938).
Note 1: The matrices in
group.data
must contain the same taxa, in the same order.Note 2: Each taxa should be present in at least 1 sample, a column with all 0's may result in errors and/or invalid results.
Value
A list containing the Xoc statistics and p-value.
References
Tvedebrink, T. (2010). Overdispersion in allelic counts and theta-correction in forensic genetics. Theor Popul Biol 78, 200-210.
Wilks, S. S. (1938). The Large-Sample Distribution of the Likelihood Ratio for Testing Composite Hypotheses. The Annals of Mathematical Statistics 9, 60-62.
Examples
data(saliva)
data(tonsils)
### Combine the data sets into a single list
group.data <- list(saliva, tonsils)
## Not run:
xoc <- Xoc.sevsample(group.data)
xoc
## End(Not run)