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
|
\name{wideByFactor}
\alias{wideByFactor}
\title{Create multivariate data by a given factor}
\description{
Modify data frame in such a way that variables are
\dQuote{separated} into several columns by factor levels.
}
\usage{
wideByFactor(x, factor, common, sort=TRUE, keepFactor=TRUE)
}
\arguments{
\item{x}{data frame}
\item{factor}{character, column name of a factor by which variables will
be divided}
\item{common}{character, column names of (common) columns that should not
be divided}
\item{sort}{logical, sort resulting data frame by factor levels}
\item{keepFactor}{logical, keep the \sQuote{factor} column}
}
\details{
Given data frame is modified so that the output represents a data
frame with \eqn{c + f + n * v} columns, where \eqn{c} is a number of common
columns for all levels of a factor, \eqn{f} is a factor column, \eqn{n} is a
number of levels in factor \eqn{f} and \eqn{v} is a number of variables that
should be divided for each level of a factor. Number of rows stays the same.
}
\value{
A data frame where divided variables have sort of \dQuote{diagonalized}
structure.
}
\author{Gregor Gorjanc}
\seealso{
\code{\link[stats]{reshape}} in the \pkg{stats} package.
}
\examples{
n <- 10
f <- 2
tmp <- data.frame(y1=rnorm(n=n),
y2=rnorm(n=n),
f1=factor(rep(letters[1:f], n/2)),
f2=factor(c(rep("M", n/2), rep("F", n/2))),
c1=1:n,
c2=2*(1:n))
wideByFactor(x=tmp, factor="f1", common=c("c1", "c2", "f2"))
wideByFactor(x=tmp, factor="f1", common=c("c1", "c2"))
}
\keyword{manip}
\keyword{misc}
|
