1\name{histBxp} 2\encoding{latin1} 3\alias{histBxp} 4\title{Plot a Histogram and a Boxplot} 5\description{ 6 Creates a histogram and a horizontal boxplot on the current graphics 7 device. 8} 9\usage{ 10histBxp(x, nclass, breaks, probability=FALSE, include.lowest=TRUE, 11 xlab = deparse(substitute(x)), 12 \dots, 13 width=0.2, boxcol=3, medcol=2, medlwd=5, whisklty=2, staplelty=1) 14} 15\arguments{ 16 \item{x}{numeric vector of data for histogram. Missing values 17 (\code{NA}s) are allowed.} 18 \item{nclass}{ 19 recommendation for the number of classes (i.e., bars) the histogram should 20 have. The default is a number proportional to the logarithm of the length 21 of \code{x}. 22 } 23 \item{breaks}{ 24 vector of the break points for the bars of the histogram. The count in the 25 i-th bar is \code{sum(breaks[i] < x <= breaks[i+1])} 26 except that if \code{include.lowest} is \code{TRUE} (the default), 27 the first bar also includes points equal to \code{breaks[1]}. If 28 omitted, evenly-spaced break points are determined from 29 \code{nclass} and the extremes of the data. 30 } 31 \item{probability}{ 32 logical flag: if \code{TRUE}, the histogram will be scaled as a probability 33 density; the sum of the bar heights times bar widths will equal \code{1}. If 34 \code{FALSE}, the heights of the bars will be counts. 35 } 36 \item{include.lowest}{ 37 If \code{TRUE} (the default), the lowest bar will include data 38 points equal to the lowest break, otherwise it will act like the 39 other bars (see the description of the \code{breaks} argument). 40 } 41 \item{xlab}{character or expression for x axis labeling.} 42 \item{\dots}{additional arguments to \code{\link{barplot}}. The 43 \code{\link{hist}} function uses the function \code{barplot} to do 44 the actual plotting; consequently, arguments to the \code{barplot} 45 function that control shading, etc., can also be given to 46 \code{hist}. See the \code{barplot} documentation for arguments 47 \code{angle}, \code{density}, \code{col}, and \code{inside}. Do not 48 use the \code{space} or \code{histo} arguments. 49 } 50 \item{width}{ 51 width of the box relative to the height of the histogram. DEFAULT is 52 \code{0.2}.} 53 \item{boxcol}{color of filled box. The default is \code{3}.} 54 \item{medcol}{ 55 the color of the median line. The special value, \code{NA}, 56 indicates the current plotting color (\code{par("col")}). The 57 default is \code{2}. If \code{boxcol=0} and \code{medcol} is not 58 explicitly specified this is set to the current plotting color 59 (\code{par("col")}). 60 } 61 \item{medlwd}{ 62 median line width. The special value \code{NA}, is used to indicate 63 the current line width (\code{par("lwd")}). The default is \code{5}. 64 } 65 \item{whisklty}{ 66 whisker line type. The special value \code{NA} indicates the 67 current line type (\code{par("lty")}). The default is \code{2} 68 (dotted line).} 69 \item{staplelty}{ 70 staple (whisker end cap) line type. The special value \code{NA} 71 indicates the current line type (\code{par("lty")}). The default is 72 \code{1} (solid line). 73 74 Graphical parameters (see \code{\link{par}}) may also 75 be supplied as arguments to this function. 76 In addition, the high-level graphics arguments described under 77 \code{par} and the arguments to \code{title} may be supplied to this 78 function.} 79} 80\details{ 81 If \code{include.lowest} is \code{FALSE} the bottom breakpoint must be 82 strictly less than the minimum of the data, otherwise (the default) it 83 must be less than or equal to the minimum of the data. The top 84 breakpoint must be greater than or equal to the maximum of the data. 85 86 This function has been called \code{hist.bxp()} for 17 years; in 2012, 87 the increasingly strong CRAN policies required a new name (which could not 88 be confused with an S3 method name). 89} 90\author{S-Plus: Markus Keller, Christian Keller; port to \R in 1990's: Martin M�chler.} 91 92\seealso{\code{\link{hist}}, \code{\link{barplot}}, 93 \code{\link{boxplot}}, \code{\link{rug}} and 94 \code{\link[Hmisc]{scat1d}} in the \CRANpkg{Hmisc} package. 95} 96\examples{ 97 lab <- "50 samples from a t distribution with 5 d.f." 98 mult.fig(2*3, main = "Hist() + Rug() and histBxp(*)") 99 for(i in 1:3) { 100 my.sample <- rt(50, 5) 101 hist(my.sample, main=lab); rug(my.sample)# for 50 obs., this is ok, too.. 102 histBxp(my.sample, main=lab) 103 } 104} 105\keyword{hplot} 106