Derives prediction rule ensembles (PREs). Largely follows the procedure for deriving PREs as described in Friedman & Popescu (2008; <doi:10.1214/07-AOAS148>), with several adjustments and improvements. The main function pre() derives a prediction rule ensemble. Functions print(), plot(), coef() and importance() can be used to inspect the generated ensemble. Function predict() generates predicted values. Functions singeplot() and pairplot() depict dependence of the output on specified predictor variables. Function cvpre() performs full cross validation of a pre to calculate the expected prediction error. Functions interact() and bsnullinteract() can be used to assess interaction effects of predictor variables.
| Version: | 0.2 |
| Depends: | R (≥ 3.1.0), partykit, glmnet, MatrixModels, Formula, methods, graphics |
| Suggests: | akima, grid, foreach |
| Published: | 2017-04-26 |
| Author: | Marjolein Fokkema [aut, cre] |
| Maintainer: | Marjolein Fokkema <m.fokkema at fsw.leidenuniv.nl> |
| BugReports: | https://github.com/marjoleinF/pre/issues |
| License: | GPL-2 | GPL-3 |
| URL: | https://github.com/marjoleinF/pre |
| NeedsCompilation: | no |
| Materials: | README NEWS |
| CRAN checks: | pre results |
| Reference manual: | pre.pdf |
| Package source: | pre_0.2.tar.gz |
| Windows binaries: | r-devel: pre_0.2.zip, r-release: pre_0.2.zip, r-oldrel: pre_0.2.zip |
| OS X El Capitan binaries: | r-release: pre_0.2.tgz |
| OS X Mavericks binaries: | r-oldrel: pre_0.2.tgz |
| Old sources: | pre archive |
Please use the canonical form https://CRAN.R-project.org/package=pre to link to this page.