[R] Boxplots using percentiles?
Jacob W. Bowers
jbowers at csm.berkeley.edu
Mon Nov 26 02:05:19 CET 2001
Hi.
I wrote a little replacement for boxplot.stats [making as few changes to
boxplot.stats as possible], and then edited boxplot.default to call
"myboxplot.stats" instead of "boxplot.stats":
#This function replaces boxplot.stats --- it uses the middle 50% and
#middle 90% instead of middle 50%(IQR) and 1.5*IQR.
myboxplot.stats <- function (x, coef = NULL, do.conf = TRUE, do.out =
TRUE)
{
nna <- !is.na(x)
n <- sum(nna)
stats <- quantile(x, c(.05,.25,.5,.75,.95), na.rm = TRUE)
iqr <- diff(stats[c(2, 4)])
out <- x < stats[1] | x > stats[5]
conf <- if (do.conf)
stats[3] + c(-1.58, 1.58) * diff(stats[c(2, 4)])/sqrt(n)
list(stats = stats, n = n, conf = conf, out = x[out & nna])
}
So, when I call boxplot(), the whiskers show the middle 90% of the
distribution.
I hope this helps.
Jake
--
Jake Bowers
Dept of Political Science
UC-Berkeley
On Sun, 25 Nov 2001, Robert Gentleman wrote:
> On Mon, Nov 26, 2001 at 11:25:36AM +1300, Nick Davis wrote:
> > The standard R boxplot appears to use quartiles to determine the height of
> > the rectangles and a range parameter - RNG - (default=1.5 I think) that
> > determines the length of the whiskers as <= RNG x Interquartile Range. Is
> > it possible to instead define the range as extending to the 95th percentile?
> > If so, how would this be done?
>
> You need to create the right kind of thing to call bxp -- which has
> some documentation but basically it is the z component that you need
> to construct.
>
> You can call boxplot with plot=FALSE to get it to do most of the work
> and then just fiddle with the parts you want to change
>
> >
> > nb, I'm plotting multiple boxplots on a single chart so a function that
> > applies to multiple series would be useful.
> >
> > Thanks,
> >
> > Nick
>
> --
> +---------------------------------------------------------------------------+
> | Robert Gentleman phone : (617) 632-5250 |
> | Associate Professor fax: (617) 632-2444 |
> | Department of Biostatistics office: M1B28
> | Harvard School of Public Health email: rgentlem at jimmy.dfci.harvard.edu |
> +---------------------------------------------------------------------------+
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> Send "info", "help", or "[un]subscribe"
> (in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch
> _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
>
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
More information about the R-help
mailing list