Package: mixgb 2.2.3

Yongshi Deng

mixgb: Multiple Imputation Through 'XGBoost'

Multiple imputation using 'XGBoost', subsampling, and predictive mean matching as described in Deng and Lumley (2024) <doi:10.1080/10618600.2023.2252501>. The package supports various types of variables, offers flexible settings, and enables saving an imputation model to impute new data. Data processing and memory usage have been optimised to speed up the imputation process.

Authors:Yongshi Deng [aut, cre], Thomas Lumley [ths]

mixgb_2.2.3.tar.gz
mixgb_2.2.3.zip(r-4.7-x86_64)mixgb_2.2.3.zip(r-4.7-arm64)mixgb_2.2.3.zip(r-4.6-x86_64)mixgb_2.2.3.zip(r-4.6-arm64)mixgb_2.2.3.zip(r-4.5-x86_64)
mixgb_2.2.3.tgz(r-4.6-x86_64)mixgb_2.2.3.tgz(r-4.6-arm64)mixgb_2.2.3.tgz(r-4.5-x86_64)mixgb_2.2.3.tgz(r-4.5-arm64)
mixgb_2.2.3.tar.gz(r-4.7-arm64)mixgb_2.2.3.tar.gz(r-4.7-x86_64)mixgb_2.2.3.tar.gz(r-4.6-arm64)mixgb_2.2.3.tar.gz(r-4.6-x86_64)
mixgb_2.2.3.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
mixgb/json (API)

# Install 'mixgb' in R:

install.packages('mixgb', repos = c('https://agnesdeng.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/agnesdeng/mixgb/issues

Uses libs:

c++– GNU Standard C++ Library v3
openmp– GCC OpenMP (GOMP) support library

Datasets:

newborn - NHANES III (1988-1994) newborn data
nhanes3 - A small subset of the NHANES III (1988-1994) newborn data

On CRAN:

cpp openmp

7.14 score 27 stars 85 scripts 376 downloads 8 exports 69 dependencies

Last updated from:d2d7b58806. Checks:15 OK. Indexed: yes.

Target	Result	Time
linux-devel-arm64	OK	153
linux-devel-x86_64	OK	171
source / vignettes	OK	254
linux-release-arm64	OK	154
linux-release-x86_64	OK	180
macos-release-arm64	OK	166
macos-release-x86_64	OK	277
macos-oldrel-arm64	OK	246
macos-oldrel-x86_64	OK	343
windows-devel-arm64	OK	404
windows-devel-x86_64	OK	250
windows-release-arm64	OK	371
windows-release-x86_64	OK	131
windows-oldrel-x86_64	OK	145
wasm-release	OK	140

Exports:check_data createNA default_params default_params_cran impute_new mixgb mixgb_cv show_var

Dependencies:backports bit bit64 boot broom cli clipr codetools cpp11 crayon data.table dplyr forcats foreach generics glmnet glue haven hms iterators jomo jsonlite lattice lifecycle lme4 magrittr MASS Matrix mice minqa mitml nlme nloptr nnet numDeriv ordinal pan pillar pkgconfig prettyunits progress purrr R6 rbibutils Rcpp RcppArmadillo RcppEigen RcppParallel Rdpack readr reformulas Rfast rlang rpart shape stringi stringr survival tibble tidyr tidyselect tzdb ucminf utf8 vctrs vroom withr xgboost zigg

Imputing newdata with a saved mixgb imputer

Impute new unseen data using a saved imputer object

Rendered fromImputing-newdata.Rmdusingknitr::rmarkdown

Last update: 2026-01-17
Started: 2022-05-18

mixgb: Multiple Imputation Through XGBoost

Introduction | Impute missing values with mixgb | Customise imputation settings | Tune hyperparameters | Inspect multiply imputed values

Rendered fromUsing-mixgb.Rmdusingknitr::rmarkdown

Last update: 2026-01-17
Started: 2022-05-18

Citation

Development and contributors

Readme and manuals

Help Manual

Help page	Topics
Sanity check for input data before imputation	check_data
Create missing values for a dataset	createNA
Auxiliary function for validating xgb.params	default_params
Auxiliary function for validating xgb.params compatible with XGBoost CRAN version	default_params_cran
Impute new data with a saved 'mixgb' imputer object	impute_new
Multiple imputation through XGBoost	mixgb
Use cross-validation to find the optimal 'nrounds'	mixgb_cv
NHANES III (1988-1994) newborn data	newborn
A small subset of the NHANES III (1988-1994) newborn data	nhanes3
Show multiply imputed values for a single variable	show_var