

   TTrreelllliiss ooff HHiissttooggrraammss

        histplot(formula, data,
        prepanel = prepanel.histplot, panel = panel.histplot,
        nbins = 5, h, x0 = -h/1000, breaks, prob = T,
        aspect = "fill", ...)

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

    formula: a formula in the form: ~ x | g1 * g2 * ...{} where
             the given variables `g1', `g2', ...{} may be omit-
             ted.  The `x' variable is a numeric response and
             is the data used to construct each histogram.
             Variables on the right hand side of the vertical
             bar are factors or shingles, and are used to par-
             cel the `x' values out to various panels.  If
             there are no given variables the initial ~ may be
             omitted.

       data: A data frame as list against which to evaluate the
             formula.

   prepanel: The prepanel function to be used.

      panel: The panel function to be used.

      nbins: The suggested number of bins. The default is
             small, as the number of data points in each panel
             is unknown at this stage.

          h: The width of the bins.

         x0: A shift for the bin sequence - the breakpoints are
             at `x0 + n * h'.

     breaks: The set of breakpoints to be used. (Usually omit-
             ted).

       prob: If `true' (the default) plot a true histogram.
             The vertical axis has a relative frequency density
             scale, so the product of the dimensions of any
             panel gives the relative frequency.  Hence the
             total area under the histogram is 1 and it is
             directly comparable with most other estimates of
             the probability density function.  If `false' plot
             the counts in the bins.

        ...: generic Trellis arguments. See under `trel-
             lis.args' for details.

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

        Plot a trellis of histograms of data.

   DDeettaaiillss::

        If `breaks' is specified, those breakpoints are used.
        Otherwise if `h' is specified, a regular grid of bins
        is used with width `h'.  If neither `breaks' nor `h' is
        specified, `nbins' is used to select a suitable `h'.
        Beware of using bandwidth selection algorithms to
        choose `h', as the data will be split, possibly
        unevenly, across panels.

        Attempts to use unequally spaced breaks with `prob=F'
        will be honoured with a warning, but this is strongly
        discouraged as misleading.

   VVaalluuee::

        an object of class `trellis', which is automatically
        plotted by `print.trellis'.

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

        `truehist', `histogram', `trellis.args'

