RfEmpImp: Multiple Imputation using Chained Random Forests

Functions for methods for multiple imputation using chained random forests. Implemented algorithms can handle missing data in both continuous and categorical variables by using prediction-based or node-based conditional distributions constructed using random forests. For prediction-based imputation, the method based on the empirical distribution of out-of-bag prediction errors of random forests and the method based on normality assumption are provided. For node-based imputation, the method based on the conditional distribution formed by predicting nodes of random forests and the method based on measures of proximities of random forests are provided. More details of the statistical methods can be found in Hong et al. (2020) <arXiv:2004.14823>.

Version: 2.0.3
Depends: R (≥ 3.5.0), mice (≥ 3.8.0), ranger (≥ 0.12.1)
Suggests: testthat (≥ 2.1.0), knitr, rmarkdown
Published: 2020-05-16
Author: Shangzhi Hong [aut, cre], Henry S. Lynn [ths]
Maintainer: Shangzhi Hong <shangzhi-hong at hotmail.com>
BugReports: https://github.com/shangzhi-hong/RfEmpImp/issues
License: GPL-3
URL: https://github.com/shangzhi-hong/RfEmpImp
NeedsCompilation: no
Citation: RfEmpImp citation info
Materials: README
CRAN checks: RfEmpImp results


Reference manual: RfEmpImp.pdf
Vignettes: intro
Package source: RfEmpImp_2.0.3.tar.gz
Windows binaries: r-devel: RfEmpImp_2.0.3.zip, r-release: RfEmpImp_2.0.3.zip, r-oldrel: RfEmpImp_2.0.3.zip
macOS binaries: r-release: RfEmpImp_2.0.3.tgz, r-oldrel: RfEmpImp_2.0.3.tgz


Please use the canonical form https://CRAN.R-project.org/package=RfEmpImp to link to this page.