asremlPlus is an R package that augments the use of ASReml-R in fitting mixed models and packages generally in exploring prediction differences. This version is compatible with both ASReml-R versions 3 and 4.1, but not 4.0.
Versions 4.x-xx of asremlPlus are a major revamp of the package and include substantial syntax changes. In particular, most functions are S3 methods and so the type of the object can be omitted from the function name when calling the function.
For more information install the package and run the R command news(package = “asremlPlus”) or consult the manual.
An overview can be obtained using ?asremlPlus. In particular, an example of its use is given towards the bottom of the help information and this is avalable as the Wheat.analysis vignette. It that shows how to select the terms to be included in a mixed model for an experiment that involves spatial variation; it also illustrates diagnostic checking and prediction production and presentation for this example. A second vignette is the Wheat.infoCriteria vignette that illustrates the facilities in asremlPlus for producing and using information criteria. Two further vignettes show how to use asremlPlus for exploring and presenting predictions from a linear mixed model analysis in the context of a three-factor factorial experiment on ladybirds: one vignette, Ladybird.asreml vignette, uses asreml and asremlPlus to produce and present predictions; the other vignette, Ladybird.lm vignette, uses lm to produce the predictions and asremlPlus to present the predictions. The vignettes can be accessed via vignette(name, package = "asremlPlus"), where name is one of "Wheat.analysis", "Wheat.infoCriteria", "Ladybird.asreml" or "Ladybird.lm".
Windows binaries and source tarballs of the latest version of asremlPlus are available for installation from my repository. Installation instructions are available there.
asremlPlus is an R package available on GitHub, so it can be installed from the RStudio console or an R command line session using the devtools command install_github. First, make sure devtools is installed, which, if you do not have it, can be done as follows:
install.packages("devtools")
Next, install asremlPlus from GitHub by entering:
devtools::install_github("briencj/asremlPlus").
Version 2.0-12 of the package is available from CRAN so that you could first install it and its dependencies using:
install.packages("asremlPlus")
If you have not previously installed asremlPlus then you could first install it and its dependencies from CRAN using:
install.packages("asremlPlus")
Otherwise, you will need to install its dependencies manually:
install.packages(c("dae", "devtools", "doParallel", "dplyr", "foreach", "ggplot2", "nloptr", "parallel", "qqplotr",
"RColorBrewer", "reshape2", "rlang", "sticky", "stringr"))
It assists in automating the selection of terms to include in mixed models when 'asreml-R' is used to fit the models. A history of the fitting of a sequence of models is kept in a data frame. Procedures are available for choosing models that conform to the hierarchy or marginality principle and for fitting and choosing between two-dimensional spatial models using correlation, natural cubic smoothing spline and P-spline models. Having obtained predictions from a linear mixed model using your favourite model fitting functions, it can also be used to compute linear functions and contrasts of predictions, to investigate prediction differences and to plot predictions. As a general rule, functions that are methods for asreml and asrtests objects require asreml-R; on the other hand, functions that are methods for alldiffs and data.frame objects do not require asreml-R.
The use of the package is exemplified in four vignettes: Wheat.analysis vignette, Wheat.infoCriteria vignette, Ladybird asreml vignette and Ladybird lm vignette. They can be accessed via vignette(name, package = "asremlPlus"), where name is one of "Wheat.analysis", "Wheat.infoCriteria", "Ladybird.asreml" or "Ladybird.lm".
The content falls into the following natural groupings:
(i) Data,
(ii) Model modification functions,
(iii) Model selection and description functions,
(iv) Model diagnostics and simulation functions,
(v) Prediction production and presentation functions,
(vi) Response transformation functions,
(vii) Object manipulation functions, and
(viii) Miscellaneous functions.
For a list of the functions for each group, see the help for asremlPlus-package or the entry in the manual for asremlPlus-package.
To use those functon in asremlPlus that are methods for asreml or asrtests objects, you must have a licensed version of the package asreml. It provides a computationally efficient algorithm for fitting mixed models using Residual Maximum Likelihood. A license can be purchased from 'VSNi' http://www.vsni.co.uk/ as asreml-R, who will supply a zip file for local installation/updating.
It also imports dae,
devtools,
doParallel, dplyr, foreach, ggplot2,
'graphics',
grDevices,
methods,
nloptr,
parallel,
qqplotr,
RColorBrewer,
reshape2,
rlang,
stats,
sticky,
stringr,
utils.
The asremlPlus package is distributed under the MIT licence -- for details see LICENSE.md.