princomp_coda: Principal Components Analysis for compositional data

In Andros-Spica/cerUB: Protocols for exploring archaeometric data

Description

Principal Components Analysis (PCA) of compositional data after applying log-ratio transformation.

Usage

princomp_coda(dt, transformation_method = "ILR", method = "robust",
  init_seed = 0, samples = 100, alr_base = 1)

Arguments

dt	Data frame containing compositional data
transformation_method	Character, the log-ratio transformation to be applied. "ALR" -> additive log-ratio, "CLR" -> centered log-ratio, "ILR" -> isometric log-ratio. Additionally, accepts "log" for applying logarithmic transformation and "std" for standardization (scaled and centred).
method	Character, "standard" for standard PCA, "robust" for robust PCA.
init_seed	Numeric, the seed for the random number generator used in best_pcaCoDa).
samples	Numeric, the number of iterations applying to samples in best_pcaCoDa) and maxiter in PCAgrid).
alr_base	Character/Numeric, the name/index of the variable to be used as divisor in additional log-ratio transformation.

Andros-Spica/cerUB documentation built on June 9, 2020, 9:22 p.m.