biokNN.impute.mi: Multiple imputation for a multilevel dataset

In biokNN: Bi-Objective k-Nearest Neighbors Imputation for Multilevel Data

Description

This function returns a list of m complete datasets, where the missing values are imputed using a bi-objective kNN method. It assumes that the class variable name is known, and the rest of the variables are numerical.

Usage

biokNN.impute.mi(
  data,
  className,
  m = 5,
  nIter = 10,
  weight = 0.5,
  k = 10,
  distance = "gower"
)

Arguments

data	A dataframe with missing values
className	name of the variable that contains the classes
m	number of imputations
nIter	number of iterations, default = 10
weight	weight of the kNN values in the objective function, default = 0.5
k	number of nearest neighbours, default = 10
distance	distance function used to get the k-nearest neighbors

Value

A dataframe with the imputed data

Examples

data(data.example)
complete.data.mi <- biokNN.impute.mi(data.example,
               className = "class",
               m = 3,
               nIter = 10,
               weight = 0.9,
               k = 15,
               distance = "gower")
# View completed data sets
str(complete.data.mi)

biokNN documentation built on April 22, 2021, 9:07 a.m.