AnthropMMD (version 4.0.3)

mmd_boot: Implementation of Fidalgo et al.'s (2022) method of bootstrap for the Mean Measure of Divergence

Description

Compute a matrix of MMD dissimilarities among bootstrapped samples of the original groups. The input data must be a “raw binary dataset”.

Usage

mmd_boot(data, angular = c("Anscombe", "Freeman"), B = 100, ...)

Value

A symmetrical dissimilarity matrix of MMD values among original groups and bootstrapped samples. This matrix is an R object of class

anthropmmd_boot.

Arguments

data

A “raw binary dataset”, as defined in the man page of start_mmd.

angular

Choice of a formula for angular transformation: either Anscombe or Freeman-Tukey transformation.

B

Numeric value: number of bootstrap samples.

...

Arguments for traits selection, passed to select_traits.

Author

Frédéric Santos,

Details

This function sticks very close to Fidalgo et al's (2022) implementation. In particular, no correction for small sample sizes is applied in the MMD formula; see Fidalgo et al's (2021) for the rationale.

Note that only a “raw binary dataset” is allowed as input, since the resampling cannot be performed properly from a table of counts and frequencies.

To get a MDS plot of the dissimilarity matrix obtained with this function, see plot.anthropmmd_boot.

References

D. Fidalgo, M. Hubbe and V. Vesolowski (2021). Population history of Brazilian south and southeast shellmound builders inferred through dental morphology. American Journal of Physical Anthropology 176(2), 192-207.

D. Fidalgo, V. Vesolowski and M. Hubbe (2022). Biological affinities of Brazilian pre-colonial coastal communities explored through boostrapped biodistances of dental non-metric traits. Journal of Archaeological Science 138, 105545.

See Also

plot.anthropmmd_boot

Examples

if (FALSE) {
## Load and visualize a raw binary dataset:
data(toyMMD)
head(toyMMD)
## Compute MMD among bootstrapped samples:
resboot <- mmd_boot(
    data = toyMMD,
    B = 50, # number of bootstrap samples
    angular = "Anscombe",
    strategy = "excludeQNPT", # strategy for trait selection
    k = 10 # minimal number of observations required per trait
)
## View part of MMD matrix among bootstrapped samples:
dim(resboot)
print(resboot[1:15, 1:15])
}