Learn R Programming
⚠️There's a newer version (2.2.3) of this package.Take me there.

mixgb (version 2.0.3)

Multiple Imputation Through 'XGBoost'

Description

Multiple imputation using 'XGBoost', subsampling, and predictive mean matching as described in Deng and Lumley (2023) . 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.

Copy Link

Version

Install

install.packages('mixgb')

Monthly Downloads

362

Version

2.0.3

License

GPL (>= 3)

Issues

4

Pull Requests

0

Stars

26

Forks

5

Maintainer

Yongshi Deng

Last Published

December 7th, 2025

Functions in mixgb (2.0.3)

nhanes3

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

Create missing values for a dataset
impute_new

Impute new data with a saved mixgb imputer object
mixgb_cv

Use cross-validation to find the optimal nrounds
default_params_cran

Auxiliary function for validating xgb.params compatible with XGBoost CRAN version
data_clean

Data cleaning
default_params

Auxiliary function for validating xgb.params
mixgb

Multiple imputation through XGBoost
%>%

Pipe operator
pmm

PMM for numeric or binary variable
mixgb-package

mixgb: Multiple Imputation Through 'XGBoost'
nhanes3_newborn

NHANES III (1988-1994) newborn data
pmm.multiclass

PMM for multiclass variable
sortNA

Sort data by increasing number of missing values
show_var

Show multiply imputed values for a single variable