two.level.multivariate.calculate.UC: Calculate Between/Within covariance matrices

In lgaborini/bayessource: Bayesian approach to evaluate whether two datasets come from the same population

Description

Two-level function: item is chosen with the third parameter.

Usage

two.level.multivariate.calculate.UC(
  population,
  idx.variables,
  idx.item,
  singular.action = "skip"
)

Arguments

population	matrix or dataframe containing data columns, item column and other columns
idx.variables	data columns (indexes or character vectors)
idx.item	item column (index or character vector)
singular.action	action to take when one of the items is observed once; one of 'skip' (default) or 'fail'

Details

Items who are observed only once cannot contribute to within covariance estimation. In that case, either we skip its contribute (default) or fail.

Multivariate functions supporting paper:

Bozza, A., Taroni, F., Marquis, R., M. Schmittbuhl - "Probabilistic evaluation of handwriting evidence: likelihood ratio for authorship"

Value

list with items:

• group.means (matrix, row*variable)

• all.means (vector)

• W (the within covariance matrix)

• B (the between covariance matrix)

See Also

Other core functions: bayessource-package, get_minimum_nw_IW(), make_priors_and_init(), marginalLikelihood_internal(), marginalLikelihood(), mcmc_postproc(), samesource_C()

lgaborini/bayessource documentation built on Nov. 9, 2021, 2:10 p.m.