Partial Least Squares (PLS) Data Analysis Methods

plsdepot contains different methods for PLS analysis of one or two data tables such as Tucker's Inter-Battery, NIPALS, SIMPLS, SIMPLS-CA, PLS Regression, and PLS Canonical Analysis. The main reference for this software is the awesome book (in French) 'La Regression PLS: Theorie et Pratique' by Michel Tenenhaus.


The R package plsdepot provides a general framework for Partial Least Squares (PLS) Data Analysis Methods. plsdepot comes with a set of functions for PLS analysis of one or two data tables such as Tucker's Inter-Battery, NIPALS, SIMPLS, SIMPLS-CA, PLS Regression, as well as PLS Canonical Analysis.

Installation

To install the stable version of plsdepot from CRAN, run in your R console:

install.packages("plsdepot")

To install the development version of plsdepot from github (using the package "devtools""), simply run in your R console:

# install.packages("devtools") 
library(devtools)
install_github('plsdepot',  username='gastonstat')

More info at http://www.gastonsanchez.com/plsdepot

Motivation

It's been more than 3 years since I first launched the R package plspm. From the feedback of several colleagues, analysts and practitioners, I believe that I did the right thing in releasing and sharing the work of my doctoral adventure. I've received a lot of emails with nice comments, compliments, and congratulations. But I've also received constructive criticism, proposals, improvement suggestions... and bug reports.

If there is something in particular that I've learned over the last 3 years, is that being the maintainer of a package can be a really tough task. And this is the main reason for me to launch plsdepot.

I've decided to split plspm in order to separate those methods specifically designed for PLS Path Modeling from the rest of the PLS methods. This makes much easier the maintainance of the functions, and it also allows me to kill the bugs in a more efficient way.

I know this decision may affect the work of some people, but in the long run is the only feasible solution that I can offer to keep my altruistic work alive.

Links

plsdepot package github

plsdepot documentation

PLS Modeling stuff

Author Contact

Gaston Sanchez (gaston.stat at gmail.com)

News

Version 0.1.17 (2012-11-12)

  • simplsca now provides all R2 coefficients for T and U components

Version 0.1.16 (2012-11-05)

NEW FEATURES AND FUNCTIONS

  • PLSTROIKA has started (restructuring my pls packages)

  • Launching the R package plsdepot

  • plsdepot is born as a split from the R package "plspm" in an effort to make the maintainance of my PLS methods much easier.

  • New implemented methods: interbat Tucker's Inter-Battery, simpls SIMPLS, simplsca SIMPLS Canonical Analysis

  • Simpler but prettier graphics have been implemented

  • Rounding of results to four decimals has been eliminated

  • documentation has been improved using new features of 'roxygen2'

  • more data sets from Michel Tenenhaus' awesome book 'La Regression PLS' have been added

BUG FIXES

  • Pain-in-the-butt warning message about 'sd() deprecated' has been tackled

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("plsdepot")

0.1.17 by Gaston Sanchez, 8 years ago


http://www.gastonsanchez.com/plsdepot, http://www.plsmodeling.com


Browse source code at https://github.com/cran/plsdepot


Authors: Gaston Sanchez


Documentation:   PDF Manual  


GPL-3 license


Suggests FactoMineR, plspm


Suggested by plsRglm.


See at CRAN