The package derives prediction rule ensembles. It implements the algorithm of Friedman & Popescu (2008), with some improvements
and adjustments. The most important improvements and adjustments are:
References1) The pre package is completely R based, allowing users better access to the results and control over the fitting process. 2) An unbiased tree induction algorithm (Hothorn, Hornik & Zeileis, 2006) is used instead of the classification and regression tree (CART) algorithm, which suffers from biased variable selection. 3) The initial ensemble of prediction rules can be generated as a bagged, boosted and/or random forest ensemble. The package is available from CRAN and GitHub. Users are adviced to use the latest GitHub version, but note that pre is under development, and much work still needs to be done. https://github.com/marjoleinF/pre Hothorn, T., Hornik, K., & Zeileis, A. (2006). Unbiased recursive partitioning: A conditional inference framework. Journal of Computational and Graphical statistics, 15(3), 651-674.Friedman, J. H., & Popescu, B. E. (2008). Predictive learning via rule ensembles. The Annals of Applied Statistics, 2(3), 916-954. |

News >