[R] boxplot statistics

[francoisromain@free.fr]

Selon bogdan romocea <br44114 at gmail.com>:

> A related comment - don't rely (too much) on boxplots. They show only
> a few things, which may be limiting in many cases and completely
> misleading in others. Here are a couple of suggestions for plots which
> you may find more useful than the standard box plots:
> 	- figure 3.27 from
> http://www.stat.auckland.ac.nz/~paul/RGraphics/chapter3.html
> 	- violin plots (see package vioplot) - density plots - histograms
> 	- box-percentile plots (bpplot from Hmisc)
> 	- quantile plots
> 	- if comparing 2 distributions, qq plots, quantile-difference plots,
> mean-difference plots etc.

Hi,

HDR (highest density regions) boxplots are interresting.
See http://addictedtor.free.fr/graphiques/RGraphGallery.php?graph=102


Romain

> > -----Original Message-----
> > From: Karin Lagesen [mailto:karin.lagesen at medisin.uio.no]
> > Sent: Tuesday, October 04, 2005 5:24 AM
> > To: r-help at r-project.org
> > Subject: [R] boxplot statistics
> >
> >
> >
> > I have read and reread the boxplot and the boxplot stats page, and I
> > still cannot understand how and what boxplot shows. I realize that
> > this might be due to me not knowing enough statistics, but anyway...
> >
> > First, how does boxplot determine the size of the box? And is the line
> > inside the box the mean or the median (or something completely
> > different?) And how does it determine how long out the whiskers should
> > go?
> >
> > Also, the boxplot.stats page talks about "hinges", what are those?
> > "The two "hinges" are versions of the first and third quartile, i.e.,
> > close to 'quantile(x, c(1,3)/4)'."
> >
> > Thankyou very much.
> >
> > Karin
> > --
> > Karin Lagesen, PhD student
> > karin.lagesen at medisin.uio.no
> > http://www.cmbn.no/rognes/