central.counts: Calculate central compositions
In provenance: Statistical Toolbox for Sedimentary Provenance Analysis

central.counts

R Documentation

Calculate central compositions

Description

Computes the logratio mean composition of a continuous mixture of point-counting data.

Usage

## S3 method for class 'counts'
central(x, ...)

Arguments

`x`	an object of class `counts`
`...`	optional arguments

Details

The central composition assumes that the observed point-counting distribution is the combination of two sources of scatter: counting uncertainty and true geological dispersion.

Value

an [5 x n] matrix with n being the number of categories and the rows containing:

theta: the ‘central’ composition.
err: the standard error for the central composition.
sigma: the overdispersion parameter, i.e. the coefficient of variation of the underlying logistic normal distribution. central computes a continuous mixture model for each component (column) separately. Covariance terms are not reported.
LL: the lower limit of a ‘1 sigma’ region for theta.
UL: the upper limit of a ‘1 sigma’ region for theta.
mswd: the mean square of the weighted deviates, a.k.a. reduced chi-square statistic.
p.value: the p-value for age homogeneity