Themplot package provides a collection of functionsdesigned for exploratory model selection.
We implement model stability and variable importance plots (Mueller andWelsh (2010);Murray,Heritier and Mueller (2013)) as well as the adaptive fence (Jiang et al. (2008);Jiang etal. (2009)) for linear and generalised linear models. We addressmany practical implementation issues with sensible defaults andinteractive graphics to highlight model selection stability. The speedof implementation comes from the leaps package and multicore support forbootstrapping.
Themplot currently only supports linear and generalisedlinear models, however work is progressing to incorporate survivalmodels and mixed models.
You can see an example of the outputhere.
Installation
Check that you’re running the most recent versions of your currentlyinstalled R packages:
update.packages()Stable release on CRAN
The mplot package has been onCRAN since June2015. You can install it from CRAN in the usual way:
install.packages("mplot")library("mplot")Development version onGithub
You can use thedevtools package to install thedevelopment version ofmplot fromGitHub:
# install.packages("devtools")devtools::install_github("garthtarr/mplot")library(mplot)Usage
A reference manual is available atgarthtarr.github.io/mplot
Citation
If you use this package to inform your model selection choices,please use the following citation:
- Tarr G, Müller S and Welsh AH (2018). “mplot: An R Package forGraphical Model Stability and Variable Selection Procedures.”Journal of Statistical Software,83(9),pp. 1–28. doi: 10.18637/jss.v083.i09.
From R you can use:
citation("mplot")toBibtex(citation("mplot"))[8]ページ先頭