lmwBootstrapTests: Performs a test on the effects from the model

In bgovaerts/LMWiRe: Linear Models for Wide Responses

Performs a test on the effects from the model

Description

Tests the significance of the effects from the model using bootstrap. This function is based on the outputs of lmwEffectMatrices. Tests on combined effects are also provided.

Usage

lmwBootstrapTests(
  resLmwEffectMatrices,
  nboot = 100,
  nCores = 2,
  verbose = FALSE
)

Arguments

resLmwEffectMatrices	A list of 12 from lmwEffectMatrices.
nboot	An integer with the number of iterations to perform.
nCores	The number of cores to use for parallel execution.
verbose	If TRUE, will display a message with the duration of execution.

Value

A list with the following elements:

f.obs

A vector of size F with the F statistics calculated on the initial data for each model term.

f.boot

A b × F matrix with the F statistics calculated for the bootstrap samples.

p.values

A vector of size F with the p-value for each model effect.

resultsTable

A 2 × F matrix with the p-value and the variance percentage for each model effect.

References

Thiel M.,Feraud B. and Govaerts B. (2017) ASCA+ and APCA+: Extensions of ASCA and APCA in the analysis of unbalanced multifactorial designs, Journal of Chemometrics

Examples

 data('UCH')
 resLmwModelMatrix <- lmwModelMatrix(UCH)
 resLmwEffectMatrices <- lmwEffectMatrices(resLmwModelMatrix = resLmwModelMatrix)

 res <- lmwBootstrapTests(resLmwEffectMatrices = resLmwEffectMatrices, nboot=10, nCores=2, verbose = TRUE)

bgovaerts/LMWiRe documentation built on Sept. 17, 2022, 12:32 a.m.