display-methods {surveillance}R Documentation

Display Methods for Surveillance Time-Series Objects

Description

Display methods for objects of class "sts".

Details

The plotting of time-series plots relies on two internal functions with plot.sts.time.one being the work-horse. Its arguments are (almost) similiar to plot.survRes. k is the column to plot.

plot.sts.time(x, type,method=x@control$name,disease=x@control$data,same.scale=TRUE,par.list=list(mfrow=magic.dim(nAreas),mar=par()$mar),...)
plot.sts.time.one(x,
  k=1,domany=FALSE,ylim=NULL,xaxis.years=TRUE, xaxis.unit=TRUE,
  xlab="time", ylab="No. infected",
  main=NULL,type="s",lty=c(1,1,2),col=c(NA,1,4), lwd=c(1,1,1), outbreak.symbol = list(pch=3,col=3, cex=1),alarm.symbol=list(pch=24, col=2, cex=1),cex=1,legend.opts=list(x="top", legend=NULL,lty=NULL,pch=NULL,col=NULL),dx.upperbound=0.5,hookFunc=function() {},...)

For spacetime plots the following internal function does all the work:

plot.sts.spacetime(x,type,legend=NULL,opts.col=NULL,labels=TRUE,wait.ms=250,cex.lab=0.7,verbose=FALSE,dev.printer=NULL,...)

print is the method for printing sts objects.

Value

The methods are called for their side-effects.

Usage

plot(x,y,type,...) print(x,...)

Arguments

x
an object of class "sts"
y
missing
type
a formula specifying the plot type, several options are possible:
observed ~ time
The observations in x are aggregated over units and the resulting univariate time-series is plotted. The plotting is done by the function plot.time.sts, which takes the same arguments as the plot.survRes function.
observed ~ time | unit
shows dim(x) plots with each showing the time-series of one observational unit. The actual plotting is done by the function plot.time.sts.one
observed ~ 1 | unit
for each unit the counts are aggregated over time and a map illustrating the counts is shown. The column names of the x@observed object are used to label the entries of the x@map. Regions with an alarm are shaded.
observed ~ 1 | unit * time
an animation consisting of nrow(x) frames is generated. Each frame contains the number of counts per region for the current row in the observed matrix. It is possible to redirect the output into files, e.g. to generate an animated GIF. See the examples.
code{alarm ~ time}
Generates a so called alarmplot for a multivariate sts object. For each time point and each series it is shown whether there is an alarm. In case of hierarchical surveillance the user can pass an additional argument lvl, which is a vector of the same length as rows in x specifying for each time series its level.
...
further arguments passed to or from other methods: in case of plotting these are passed to plot, in case of printing these are passed to print.default

See Also

plot.survRes

Examples

data(ha)
shp <- system.file("shapes/berlin.shp",package="surveillance")
has4 <- disProg2sts(ha, map=readShapePoly(shp,IDvar="SNAME"))

print(has4)
plot(has4, type= observed ~ time)
plot(has4, type= observed ~ time | unit)
plot(has4, type= observed ~ 1 | unit)
plot(has4[1:20,1:2], type= observed ~ 1 | unit)
plot(aggregate(has4,nfreq=13), type= observed ~ 1 | unit * time)

## Not run: 
#Configure a png device printer for the plot command
dev.printer <-
list(device=png,extension=".png",width=640,height=480,name="/tmp/berlin")

#Do the animation
plot(aggregate(has4,nfreq=13), type = observed ~ 1 | unit * time, dev.printer=dev.printer)

#Use ImageMagick -- replace /sw/bin/convert by your path to convert
system(paste("/sw/bin/convert -delay 50 ",dev.printer$name,"*.png /tmp/animated.gif",sep=""))
## End(Not run)


[Package surveillance version 1.0-3 Index]