1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
|
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/kNN.R
\name{kNN}
\alias{kNN}
\title{k-Nearest Neighbour Imputation}
\usage{
kNN(
data,
variable = colnames(data),
metric = NULL,
k = 5,
dist_var = colnames(data),
weights = NULL,
numFun = median,
catFun = maxCat,
makeNA = NULL,
NAcond = NULL,
impNA = TRUE,
donorcond = NULL,
mixed = vector(),
mixed.constant = NULL,
trace = FALSE,
imp_var = TRUE,
imp_suffix = "imp",
addRF = FALSE,
onlyRF = FALSE,
addRandom = FALSE,
useImputedDist = TRUE,
weightDist = FALSE,
methodStand = "range",
ordFun = medianSamp
)
}
\arguments{
\item{data}{data.frame or matrix}
\item{variable}{variables where missing values should be imputed}
\item{metric}{metric to be used for calculating the distances between}
\item{k}{number of Nearest Neighbours used}
\item{dist_var}{names or variables to be used for distance calculation}
\item{weights}{weights for the variables for distance calculation.
If \code{weights = "auto"} weights will be selected based on variable importance from random forest regression, using function \code{\link[ranger:ranger]{ranger::ranger()}}.
Weights are calculated for each variable seperately.}
\item{numFun}{function for aggregating the k Nearest Neighbours in the case
of a numerical variable}
\item{catFun}{function for aggregating the k Nearest Neighbours in the case
of a categorical variable}
\item{makeNA}{list of length equal to the number of variables, with values, that should be converted to NA for each variable}
\item{NAcond}{list of length equal to the number of variables, with a condition for imputing a NA}
\item{impNA}{TRUE/FALSE whether NA should be imputed}
\item{donorcond}{list of length equal to the number of variables, with a donorcond condition as character string.
e.g. a list element can be ">5" or c(">5","<10). If the list element for a variable is NULL no condition will be applied for this variable.}
\item{mixed}{names of mixed variables}
\item{mixed.constant}{vector with length equal to the number of
semi-continuous variables specifying the point of the semi-continuous
distribution with non-zero probability}
\item{trace}{TRUE/FALSE if additional information about the imputation
process should be printed}
\item{imp_var}{TRUE/FALSE if a TRUE/FALSE variables for each imputed
variable should be created show the imputation status}
\item{imp_suffix}{suffix for the TRUE/FALSE variables showing the imputation
status}
\item{addRF}{TRUE/FALSE each variable will be modelled using random forest regression (\code{\link[ranger:ranger]{ranger::ranger()}}) and used as additional distance variable.}
\item{onlyRF}{TRUE/FALSE if TRUE only additional distance variables created from random forest regression will be used as distance variables.}
\item{addRandom}{TRUE/FALSE if an additional random variable should be added
for distance calculation}
\item{useImputedDist}{TRUE/FALSE if an imputed value should be used for distance calculation for imputing another variable.
Be aware that this results in a dependency on the ordering of the variables.}
\item{weightDist}{TRUE/FALSE if the distances of the k nearest neighbours should be used as weights in the
aggregation step}
\item{methodStand}{either "range" or "iqr" to be used in the standardization of numeric vaiables in the gower distance}
\item{ordFun}{function for aggregating the k Nearest Neighbours in the case
of a ordered factor variable}
}
\value{
the imputed data set.
}
\description{
k-Nearest Neighbour Imputation based on a variation of the Gower Distance
for numerical, categorical, ordered and semi-continous variables.
}
\examples{
data(sleep)
kNN(sleep)
library(laeken)
kNN(sleep, numFun = weightedMean, weightDist=TRUE)
}
\references{
A. Kowarik, M. Templ (2016) Imputation with
R package VIM. \emph{Journal of
Statistical Software}, 74(7), 1-16.
}
\seealso{
Other imputation methods:
\code{\link{hotdeck}()},
\code{\link{impPCA}()},
\code{\link{irmi}()},
\code{\link{matchImpute}()},
\code{\link{medianSamp}()},
\code{\link{rangerImpute}()},
\code{\link{regressionImp}()},
\code{\link{sampleCat}()}
}
\author{
Alexander Kowarik, Statistik Austria
}
\concept{imputation methods}
\keyword{manip}
|
