palaeoSig (version 2.1-4)

multi.mat: MAT for multiple variables

Description

MAT for many environmental variables simultaneously. More efficient than calculating them separately for each variable.

Usage

multi.mat(
  training.spp,
  envs,
  core.spp,
  noanalogues = 10,
  method = "sq-chord",
  run = "both"
)

Value

If run = "both", a list with two elements:

jack

Matrix of leave-one-out cross-validation predictions for the calibration set

core

Matrix of predictions for the fossil data

Otherwise, one of these matrices is returned.

Arguments

training.spp

Community data

envs

Environmental variables - or simulations

core.spp

Optional fossil data to make predictions for

noanalogues

Number of analogues to use

method

distance metric to use

run

Return LOO predictions or predictions for fossil data

Author

Richard Telford

References

Telford, R. J. and Birks, H. J. B. (2009) Evaluation of transfer functions in spatially structured environments. Quaternary Science Reviews 28: 1309--1316. tools:::Rd_expr_doi("10.1016/j.quascirev.2008.12.020")

Examples

data(arctic.env)
data(arctic.pollen)

mMAT <- multi.mat(arctic.pollen, arctic.env[, 9:67], noanalogues = 5)