here is the hidden area text

Introduction

The GAMLSS software is implemented in a series of packages in the R language ((R Development Core Team, 2013), and it is available from CRAN the R library at http://www.r-project.org.

The themes in this page are

  • Packages
  • Manual and other help for the packages
  • Downloading the extra R GAMLSS packages
  • ACEGES and GAMLSS
  • Third party GAMLSS packages for R
Packages

The GAMLSS software currently comprises of the following different packages:

* the original gamlss package for fitting GAMLSS (now depends on gamlss.dist and gamlss.data
* the gamlss.add package for extra additive terms.
* the gamlss.boot experimental package for bootstrapping centile curves (not in CRAN).
* the gamlss.cens package for fitting censored (interval) response variables.
* the gamlss.data package for all example data used in GAMLSS.
* the gamlss.demo package for demos using the R package rpanel.
* the gamlss.dist package for all gamlss.family distributions.
* the gamlss.mx package for fitting finite mixture distributions.
* the gamlss.nl package for fitting nonlinear models.
* the gamlss.rsm experimental package for fitting randomly stopped models (not in CRAN).
* the gamlss.spatial package for fitting  Gaussian Markov random fields (MRF) models in GAMLSS.
* the gamlss.tr package for fitting truncated distributions.
* the gamlss.util package having functions not necessarily related to GAMLSS.

All the gamlss.family distributions are now implemented in the package gamlss.dist .

Note that dependencies of the packages have changed radically in version 3.0-0 and also for version 4.2-0.

Manuals and other help for the packages

The gamlss packages reference card is available here

Each package has its individual help files. Here is a list of what users may find useful.

  • The second edition of the manual of the gamlss package in pdf form. This is a good starting point covering most of the topics of the original gamlss package but it is now dated since it was created in 2008.
  • The Journal of Statistical Software which has a brief introduction to GAMLSS and shows how the models can be used in practice.
  • Course notes from the Lancaster 2009 short course.
Downloading the extra R GAMLSS packages

The following table shows packages which are not currently in CRAN :

The current GAMLSS packages

the packages zip for windows the tar zip files
gamlss.boot 1.6.5 (test version) zip tar
gamlss.rms 1.0.0 (test version of randomly stopped models) zip tar
gamlss.sparse 0.0.1 (using sparse matrices) zip tar

ACEGES and GAMLSS

The ACEGES decision-support tools is an agent-based model for exploratory energy policy by means of controlled computational experiments. The ACEGES tool is designed to be the foundation for large custom-purpose simulations of the global energy system. GAMLSS is the back-end statistical model for the regression-based rules of the agents in ACEGES.

Webpage: ACEGES
Related papers: Voudouris (2011) and Jefferson and Voudouris (2011).

Third party GAMLSS packages for R

gamboostLSS – Boosting Methods for GAMLSS models.
Summary: The package provides boosting methods for fitting generalized additive models for location, scale and shape (GAMLSS) to potentially high dimensional data.
Authors: Benjamin Hofner ( email), Andreas Mayr ( email), Nora Fenske ( email) and Matthias Schmid ( email).
R-Forge page: http://r-forge.r-project.org/projects/gamboostlss/
Related paper: http://epub.ub.uni-muenchen.de/11938/1/TR098.pdf