ensembleTest: Tests the correlation across a Bootstrap ensemble to ensure a...

In jeffreyevans/rfUtilities: Random Forests Model Selection and Performance Evaluation

Tests the correlation across a Bootstrap ensemble to ensure a quasi-independent sample and uncorrelated ensemble.

Description

Tests the correlation across a Bootstrap ensemble to ensure a quasi-independent sample and uncorrelated ensemble.

Usage

ensembleTest(
  x,
  perm = 99,
  p = 0.05,
  sampling = c("boot", "elimination"),
  type = c("mvc", "cov", "psi", "pairwise"),
  n = NULL,
  replacement = TRUE
)

Arguments

x	A matrix or data.frame of the data to test
perm	Number of Bootstraps (eg., same as random forests ntree)
p	Accept/Reject p-value
type	Type of matrix similarity statistic
n	Sample proportion
replacement	(TRUE/FALSE) Sample with replacement

Note

type options are:

• mcv - multivariate correlation (RV-coefficient)

• cov - covariance equivalence

• psi - Procrustes Similarity Index

• pairwise - Averaged column pairwise comparison

Author(s)

Jeffrey S. Evans <jeffrey_evans<at>tnc.org>

References

Robert, P., Escoufier, Y. (1976). A Unifying Tool for Linear Multivariate Statistical Methods: The RV-Coefficient. Applied Statistics 25(3):257-265.

Smilde, AK., Kiers, HA., Bijlsma, S., Rubingh, CM., van Erk, MJ (2009) Matrix correlations for high-dimensional data: the modified RV-coefficient. Bioinformatics 25(3): 401-5.

library(raster) r <- stack("C:/evans/India/Chennai/data/2019_OLI8.tif") x <- sampleRandom(r, 5000)

jeffreyevans/rfUtilities documentation built on Nov. 12, 2023, 6:52 p.m.