mixgb: Multiple Imputation Through 'XGBoost'

Multiple imputation using 'XGBoost', subsampling, and predictive mean matching as described in Deng and Lumley (2023) <arXiv:2106.01574>. Our method utilizes the capabilities of XGBoost, a highly efficient implementation of gradient boosted trees, to capture interactions and non-linear relations automatically. Moreover, we have integrated subsampling and predictive mean matching to minimize bias and reflect appropriate imputation variability. This package supports various types of variables and offers flexible settings for subsampling and predictive mean matching. Additionally, it includes diagnostic tools for evaluating the quality of the imputed values.

Version: 1.0.2
Depends: R (≥ 3.5.0)
Imports: data.table, ggplot2, Matrix, mice, Rfast, rlang, scales, stats, tidyr, utils, xgboost
Suggests: knitr, rmarkdown, RColorBrewer
Published: 2023-02-16
Author: Yongshi Deng ORCID iD [aut, cre], Thomas Lumley [ths]
Maintainer: Yongshi Deng <yongshi.deng at auckland.ac.nz>
BugReports: https://github.com/agnesdeng/mixgb/issues
License: GPL (≥ 3)
URL: https://github.com/agnesdeng/mixgb, https://agnesdeng.github.io/mixgb/
NeedsCompilation: no
CRAN checks: mixgb results

Documentation:

Reference manual: mixgb.pdf
Vignettes: Imputing newdata with a saved mixgb imputer
mixgb: Multiple Imputation Through XGBoost

Downloads:

Package source: mixgb_1.0.2.tar.gz
Windows binaries: r-devel: mixgb_1.0.2.zip, r-release: mixgb_1.0.2.zip, r-oldrel: mixgb_1.0.2.zip
macOS binaries: r-release (arm64): mixgb_1.0.2.tgz, r-oldrel (arm64): mixgb_1.0.2.tgz, r-release (x86_64): mixgb_1.0.2.tgz
Old sources: mixgb archive

Linking:

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