

   bbooxxpplloott

        bplot(x, ..., xpos=NA, width, label, by, srt=0, add=F, space=0.25, sort.names=T, xlab="", ylab="")

   AArrgguummeennttss::

          x: Vector, matrix,  list or data frame

        ...: Additional data sets to be plotted.

       xpos: The boxes will be plotted vertically and xpos
             gives the x locations for their centers. If omit-
             ted the boxes are equally spaced at integer val-
             ues.

      width: Width of boxplots if omitted then the width is a
             reasonable fraction of the distance between boxes
             and is set by the space argument.

      label: Labels under each boxplot. If missing the columns
             names or components of x are used.

         by: If x is a vector a vector of categories that will
             be used to split the x values into separate
             groups. Of course box plots will be made of the
             groups.

        srt: Rotate the labels. Default is to put them horizon-
             tal.

        add: Do not create a new plots just add the boxplots to
             a current plot. Note that the xpos argument may be
             useful in this case.

      space: Space between boxplots.

   sort.names: Plot the boxplots sorted in alphabetic order by
             their labels.

       xlab: Label for the x-axis

       ylab: Label for the y-axis

   DDeessccrriippttiioonn::

        This function was created as a complement to the usual
        S function for boxplots. The current function makes it
        possible to put the boxplots at unequal x-positions.
        This is useful for visually grouping a large set of
        boxplots into several groups. Also placement of the
        boxplots with respect to the x-axis can add information
        to the plot.  Another aspect is the emphasis on data
        structures for groups of data. One useful feature is
        the by option to break up the x vector into distinct
        groups.  If 5 or less observations are in a group the
        points themselves are plotted instead of a box.

        Finally it should be noted that this function con-
        structs the boxplots based on percentiles of the data
        rather than using the hinge and fence arcana from EDA.
        This is probably heretical to some but it makes it much
        easier to  teach undergraduate students what these are!

        The function is simple enough so that one _could_
        hingeify the boxes.  Also it would be simple to modify
        this function to get boxes horizontally instead of ver-
        tically.

   SSeeee AAllssoo::

        boxplot,lplot

   EExxaammpplleess::

        # boxplots of the times for different events in a minitriathlon
        bplot( minitri)
        # some boxplots of uniform random sample with different power
        # transformations
        temp<- runif(50)
        bplot( temp, temp**1.5, temp**2, xpos=c(1,1.5,2.0), labels=c("uniform","1.5
        power", "square") )

