invSLR: Inverse of full set of amalgamation balances

In easyCODA: Compositional Data Analysis in Practice

Description

Given a full set of amalgamation (or summation) balances (SLRs), compute the inverse (i.e. recover the original parts)

Usage

invSLR(SLRmatrix, part.names=NA, ratio.names=colnames(SLRmatrix))

Arguments

SLRmatrix	A matrix of amalgamation logratios, one less column than the number of parts
part.names	Part names in the reconstructed compositional data matrix
ratio.names	Definition of the amalgamation logratios

Details

The function invSLR computes the original parts, given the amalgamation logratios (CLRs). The amalgamation logratios are specified in ratio.names in the format num/den where num and den are the numerator and denominator amalgamations respectively. An amalgamation is specified as "p1&p2&...", where p1, p2, etc. are the parts summed in the amalgamation. For example, an SLR of the ratio MnO/(CaO+P2O5) would be names as "MnO/CaO&P2O5".

Value

parts

The reconstructed parts (they add up to 1)

Author(s)

Michael Greenacre

References

Aitchison, J. (1986), The Statistical Analysis of Compositional Data, Chapman & Hall.
Greenacre, M. (2018), Compositional Data Analysis in Practice, Chapman & Hall / CRC Press.

See Also

CLR, ALR, invALR, LR.VAR

Examples

data(veg)
# compute centred logratios
veg.CLR <- CLR(veg)$LR
# invert back to original parts (parts closed to sum to 1)
invALR(veg.CLR)

easyCODA documentation built on Sept. 20, 2020, 1:07 a.m.