volins {caroline} | R Documentation |
This is an improvment on the original vioplot function (combination of a box plot and a kernel density plot). You can now pass a list of violins and a list of colors. Optionally it also returns statistics on the distributions as well.
violins(x, by, range = 1.5, h = NULL, ylim = NULL, names = NULL, horizontal = FALSE, col = "transparent", border = "black", lty = 1, lwd = 1, rectCol = "grey50", colMed = "grey80", pchMed = 19, at, add = FALSE, wex = 1, drawRect = TRUE, main = "", xlab = "", ylab = "", connect = c("median", "mean", "hubermu", "deciles"), SD.or.SE = c("SD"), connectcol = c("lightblue", "cyan", "darkred", "grey"), las = 2, stats = FALSE, quantiles = c(0.1, 0.9), CImed = TRUE, deciles = TRUE)
x |
list of vectors or a dataframe |
by |
accepts 'by' object for grouping |
range |
a factor to calculate the upper/lower adjacent values. |
h |
the height for the density estimator, if omit as explained in sm.density, h will be set to an optimum. |
ylim |
y limits. |
names |
one label, or a vector of labels for the datas must match the number of datas given. |
horizontal |
description of var 1. |
col,border,lty,lwd |
Graphical parameters for the violin passed to lines and polygon. |
rectCol,colMed,pchMed |
Graphical parameters to control the look of the box. |
at |
position of each violin. Default to 1:n |
add |
logical. if FALSE (default) a new plot is created |
wex |
relative expansion of the violin. |
drawRect |
logical. the box is drawn if TRUE. |
main |
main title for the plot. |
connect |
connects a violin plot series at the medians with line segments |
connectcol |
line color of the median connecting line segments |
xlab |
x axis label |
ylab |
y axis label |
stats |
print out statistics |
SD.or.SE |
Boolean for standard deviation or standard error |
las |
axis tick mark labels orientation |
quantiles |
map any pair of quantiles (as dotted box) in addition to Q1 & Q3, but are not shown when "c(0,0)" and arg is passed to descriptive stats when 'stats'=TRUE |
CImed |
portrays 95 percent confidence intervals for the median (as solid box) |
deciles |
maps deciles 0.1:0.9 (as thin lines) independently of 'quantiles' and can be connected when 2 or more variables are plotted. |
a series of violin plots
Daniel Adler <dadler@uni-goettingen.de>, David Schruth, Dave McArthur <dmca@ucla.edu>
vioplot(vioplot), sm.density
n <- rnorm(130, 10, 3) p <- rpois(110, 4) u <- runif(300, 0, 20) l <- rlnorm(130, log(2)) g <- rgamma(140, 3) e <- rexp(160) violins(list(e=e, p=p,u=u,n=n,l=l,g=g), ylim=c(0,20), col=c('purple','lightblue','lightgreen','red','orange','yellow'), stats=TRUE)