- x
an object of class "epidataCS".
- aggregate
character, one of "time" and "space", referring to the
specific plot functions epidataCSplot_time and
epidataCSplot_time, respectively.
For "time", the number of events over time is plotted as
hist (or hist.Date).
For "space", the observation region x$W (or the
tiles) and the locations of the events therein are plotted.
- subset
logical expression indicating a subset of events to consider for
plotting: missing values are taken as false. Note that the
expression is evaluated in the data frame of event marks
(marks(x)), which means that column names can be referred to
by name (like in subset.data.frame).
- ...
in the basic plot-method further arguments are passed to the
aggregate-specific plot function.
In epidataCSplot_time, further graphical parameters are
passed to hist or hist.Date,
respectively. In epidataCSplot_space, further arguments are
passed to the plot-method for
"SpatialPolygons", which draws tiles.
- by
an expression evaluated in marks(x), defining how
events should be stratified in the plot (the result is converted to
a factor), or NULL to disregard event types.
By default (by = type) the plot distinguishes between event
types, i.e., the bars of the temporal plot are stacked by type, and
the point colors in the spatial plot differ by type, respectively.
Note: to select specific event types for plotting use the
subset argument, e.g., subset=(type=="B").
- t0.Date
the beginning of the observation period
t0 = x$stgrid$start[1] as a "Date" (or
anything coercible by as.Date without further arguments),
enabling a nice x-axis using hist.Date and sensible
breaks of the histogram, e.g., breaks="months".
The event times then equal
t0.Date + as.integer(x$events$time - t0), i.e. possible
fractional parts of the event times are removed (which ensures that
using breaks = "months" or other automatic types always
works).
- breaks
a specification of the histogram break points, see
hist (or hist.Date if t0.Date is
used). The default value "stgrid" is special and means to use
the temporal grid points
with(x$stgrid, c(start[1L], unique.default(stop)))
as breaks (or their "Date" equivalents).
- freq
see hist, defaults to TRUE.
- col
fill colour for the bars of the histogram, defaults to
the vector of rainbow colours.
- cumulative
if a list (of style options),
lines for the cumulative number of events (per type) will be
added to the plot. Possible options are axis (logical),
lab (axis label), maxat (single integer affecting
the axis range), lwd, col, and offset (a
numeric vector of length the number of types).
- add
logical (default: FALSE) indicating if the plot
should be added to an existing window.
Ignored if an spplot is created (if pop is
non-NULL).
- mar
see par. The default (NULL) is
mar <- par("mar"), with mar[4] <- mar[2] if an
axis is requested for the cumulative numbers.
- xlim,ylim
NULL provides automatic axis limits.
- xlab,ylab
axis labels (with sensible defaults).
- main
main title of the plot (defaults to no title).
- panel.first
expression that should be evaluated after the
plotting window has been set up but before the histogram is plotted.
Defaults to adding horizontal grid lines.
- legend.types
if a list (of arguments for legend),
a legend for the event types is added to the plot in case there is
more than one type.
- tiles
the observation region x$W (default) or,
alternatively, a "SpatialPolygons"
representation of the tiles of x$stgrid.
- pop
if tiles is a
"SpatialPolygonsDataFrame", pop can
specify an attribute to be displayed in a levelplot behind the
point pattern, see spplot. By default (NULL),
the conventional graphics system is used to display the tiles
and event locations, otherwise the result is a
trellis.object.
- cex.fun
function which takes a vector of counts of events
at each unique location and returns a (vector of) cex
value(s) for the sizes of the corresponding points.
Defaults to the sqrt() function, which for the default
circular pch=1 means that the area of each point is
proportional to the number of events at its location.
- points.args
a list of (type-specific) graphical parameters
for points, specifically pch, lwd,
and col, which are all recycled to give the length
nlevels(x$events$type). In contrast, a possible
cex element should be scalar (default: 0.5) and
multiplies the sizes obtained from cex.fun.
- legend.counts
if a list (of arguments for
legend), a legend illustrating the effect of
cex.fun is added to the plot. This list may contain a
special element counts, which is an integer vector
specifying the counts to illustrate.
- sp.layout
optional list of additional layout items in case
pop is non-NULL, see spplot.