Usage
kaplan.plot(x="TIME",y="DV",id="ID",
data=NULL,
evid="EVID",
by=NULL,
xlab="Time",
ylab="Default",
object=NULL,
events.to.plot="All",
sim.data=NULL,
sim.zip.file=NULL,
VPC = FALSE,
nsim.lab="simNumber",
sim.evct.lab="counter",
probs=c(0.025,0.975),
add.baseline=T,
add.last.area=T,
subset=NULL,
##subset.real=NULL,
main="Default",
main.sub="Default",
main.sub.cex = 0.8,
nbins=NULL,
real.type="l",
real.lty=1,
real.lwd=1,
real.col="blue",
real.se= if(!is.null(sim.data)) F else T,
real.se.type="l",
real.se.lty=2,
real.se.lwd=0.5,
real.se.col="red",
cens.type="l",
cens.lty=1,
cens.col="black",
cens.lwd=1,
inclZeroWRES=TRUE,
onlyfirst=FALSE,
samp=NULL,
poly.alpha=1,
poly.fill="lightgreen",
poly.line.col="darkgreen",
poly.lty=2,
censor.lines=TRUE,
ylim=c(-5,105),
cov = NULL,
cov.fun = "mean",
...)Arguments
x
The indenpendent variable.
y
The dependent variable. event (>0) or no event (0).
id
The ID variable in the dataset.
data
A dataset can be used instead of the data in an Xpose object. Must have
the same form as an xpose data object xpdb@Data.
evid
The EVID data item. If not present then all rows are considered events
(can be censored or an event). Otherwise, EVID!=0 are dropped from the data set.
by
A vector of conditioning variables.
object
An Xpose object. Needed if no data is supplied.
events.to.plot
Vector of events to be plotted. "All" means that all events are plotted.
sim.data
The simulated data file. Should be a table file with one header row
and have, at least, columns with headers coeesponding to
x, y, id, by (if used), nsim.lab
and sim.evct.lab
sim.zip.file
The sim.data can be in .zip format and xpose will unzip the file
before reading in the data. Must have the same structure as described
above in sim.data.
VPC
TRUE or FALSE. If TRUE then Xpose will search for
a zipped file with name
paste("simtab",object@Runno,".zip",sep=""), for example "simtab42.zip".
nsim.lab
The column header for sim.data that contains the simulation
number for that row in the data.
sim.evct.lab
The column header for sim.data that contains the individual event counter
information. For each individual the event counter should increase by
one for each event (or censored event) that occurs.
probs
The probabilities (non-parametric percentiles) to use in computation
of the prediciton intervals for the simulated dats.
add.baseline
Should a (x=0,y=1) baseline measurement be added to each individual in the
dataset. Otherwise each plot will begin at the first event in the dataset.
add.last.area
Should an area be added to the VPC extending the last PI?
subset
The subset of the data and sim.data to use.
main
The title of the plot. Can also be NULL or "Default".
main.sub
The title of the subplots. Must be a list, the same length as the number
of subplots (actual graphs), or NULL or "Default".
main.sub.cex
The size of the title of the subplots.
nbins
The number of bins to use in the VPC. If NULL, the the number of
unique x values in sim.data is used.
real.se
Should the standard errors of the real (non simulated) data be plotted?
Calculated using survfit. real.se.type
Type for the standard errors.
real.type
Type for the real data.
real.lwd
Line width (lwd) for the real data.
real.lty
Line type (lty) for the curve of the original (or real) data.
real.col
Color for the curve of the original (or real) data.
real.se.lty
Line type (lty) for the standard error lines.
real.se.lwd
Line width (lwd) for the standard error lines.
real.se.col
Color for the standard error lines.
cens.type
Type for the censored lines.
cens.lty
Line type (lty) for the censored lines.
cens.col
Color for the censored lines.
cens.lwd
Line width for the censored lines.
cov
The covariate in the dataset to plot instead of the survival curve.
cov.fun
The summary function for the covariate in the dataset to plot instead of the survival curve.
inclZeroWRES
Include WRES=0 rows from the real data set in the plots?
onlyfirst
Include only the first measurement for the real data in the plots?
samp
Simulated data in the xpose data object can be used as the "real" data.
samp is a number selecting which simulated data set to use.
poly.alpha
The tranparency of the VPC shaded region.
poly.fill
The fill color of the VPC shaded region.
poly.line.col
The line colors for the VPC region.
poly.lty
The line type for the VPC region.
censor.lines
Should censored observations be marked on the plot?
ylim
Limits for the y-axes
...
Additional arguments passed to the function.