Power Analysis for PLS Classification

It estimates power and sample size for Partial Least Squares-based methods described in Andreella, et al., (2024), .

powerPLS

This package provides a tool to perform power analysis in Partial Least Squares (PLS) for classification when two classes are analyzed.

Installation

You can install the released version of powerPLS with:

devtools::install_github("angeella/powerPLS")

Quick overview

The main functions are

• computeSampleSize() which estimated the power considering several values of sample size and number of score components.

• computePower() which estimated the power considering a fixed sample size and several number of score components.

datas <- simulatePilotData(nvar = 10, clus.size = c(5,5),m = 6,nvar_rel = 5,A = 2)
out <- computePower(X = datas$X, Y = datas$Y, A = 3, n = 20, test = "R2")
out <- computeSampleSize(X = datas$X, Y = datas$Y, A = 2, A = 3, n = 20, test = "R2")

References

Andreella, A., Finos, L., Scarpa, B. and Stocchero, M. "Towards a power analysis for PLS-based methods" arXiv:2403.10289 stat.ME. link: https://arxiv.org/abs/2403.10289

Did you find some bugs?

Please write to angela.andreella[\at]unitn[\dot]it or insert a reproducible example using reprex on my issue github page.