miss_sim: Simulation of missings

In ClustImpute: K-Means Clustering with Build-in Missing Data Imputation

Description

Simulates missing at random using a normal copula to create correlations between the missing (type="MAR"). Missings appear in each column of the provided data frame with the same ratio.

Usage

miss_sim(dat, p = 0.2, type = "MAR", seed_nr = 123)

Arguments

dat	Data frame with only numeric values
p	Fraction of missings (for entire data frame)
type	Type of missingness. Either MCAR (=missing completely at random) or MAR (=missing at random)
seed_nr	Number for set.seed()

Value

data frame with only numeric values and NAs

Examples

data(cars)
cars_with_missings <- miss_sim(cars,p = .2,seed_nr = 4)
summary(cars_with_missings)

ClustImpute documentation built on May 31, 2021, 9:06 a.m.