mfix: Four single imputation methods

In roughrf: Roughened Random Forests for Binary Classification

Description

Four single imputation methods can be implemented by this function, including median/mode imputation (median imputation on continuous variables and mode imputation on categorical variables), mean/mode imputation (mean imputation on continuous variables and mode imputation on categorical variables), minimum-value/mode imputation (minimum-value imputation on continuous variables and mode imputation on categorical variables) and maximum-value/mode imputation (maximum-value imputation on continuous variables and mode imputation on categorical variables). When mmmm=1, this function is the same as na.roughfix from R package randomForest.

Usage

mfix(x, mmmm)

Arguments

x	A dataset with missing values.
mmmm	Its value is used to select from four different single imputation methods. mmmm=1 refers to median/mode imputation; mmmm=2 refers to mean/mode imputation; mmmm=3 refers to minimum-value/mode imputation; mmmm=4 refers to maximum-value/mode imputation.

Value

Imputed dataset

Author(s)

Kuangnan Xiong

References

Liaw, A. & Wiener, M., 2002. Classification and regression by randomForest. R News, 2(3), pp. 18-22.

See Also

rrfa, rrfb, rrfc1, rrfc2, rrfc3, rrfc4,rrfc5, rrfc6, rrfc7, rrfd, rrfe

Examples

dat=data.frame(continuous=c(1,2,3,4,5),categorical=c('a','a','a','a','b'))
dat[2,]=NA
dat
summary(dat)
mfix(dat,1)[2,]  #median/mode imputation
mfix(dat,2)[2,]  #mean/mode imputation
mfix(dat,3)[2,]  #minimum-value/mode imputation
mfix(dat,4)[2,]  #maximum-value/mode imputation
#

roughrf documentation built on May 2, 2019, 9:30 a.m.