Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!
Get Started for Free

plot.Hist

Box-arrow diagrams for multi-state models.

Description

Automated plotting of the states and transitions that characterize a multi states model.

Usage

## S3 method for class 'Hist'
plot(x, nrow, ncol, stateLabels, arrowLabels,
  arrowLabelStyle = "symbolic", arrowLabelSymbol = "lambda",
  changeArrowLabelSide, tagBoxes = FALSE, startCountZero = TRUE,
  oneFitsAll, margin, cex, verbose = FALSE, ...)

Arguments

x

An object of class Hist.
nrow

the number of graphic rows
ncol

the number of graphic columns
stateLabels

Vector of names to appear in the boxes (states). Defaults to attr(x,"state.names"). The boxes can also be individually labeled by smart arguments of the form box3.label="diseased", see examples.
arrowLabels

Vector of labels to appear in the boxes (states). One for each arrow. The arrows can also be individually labeled by smart arguments of the form arrow1.label=paste(expression(eta(s,u))), see examples.
arrowLabelStyle

Either "symbolic" for automated symbolic arrow labels, or "count" for arrow labels that reflect the number of transitions in the data.
arrowLabelSymbol

Symbol for automated symbolic arrow labels. Defaults to "lambda".
changeArrowLabelSide

A vector of mode logical (TRUE,FALSE) one for each arrow to change the side of the arrow on which the label is placed.
tagBoxes

Logical. If TRUE the boxes are numbered in the upper left corner. The size can be controlled with smart argument boxtags.cex. The default is boxtags.cex=1.28.
startCountZero

Control states numbers for symbolic arrow labels and box tags.
oneFitsAll

If FALSE then boxes have individual size, depending on the size of the label, otherwise all boxes have the same size dependent on the largest label.
margin

Set the figure margin via par(mar=margin). Less than 4 values are repeated.
cex

Initial cex value for the state and the arrow labels.
verbose

If TRUE echo various things.
...

Smart control of arguments for the subroutines text (box label), rect (box), arrows, text (arrow label). Thus the three dots can be used to draw individual boxes with individual labels, arrows and arrow labels. E.g. arrow2.label="any label" changes the label of the second arrow. See examples.

Note

Use the functionality of the unix program ‘dot’ http://www.graphviz.org/About.php via R package Rgraphviz to obtain more complex graphs.

Author(s)

Thomas A Gerds tag@biostat.ku.dk

See Also

Examples

## A simple survival model

SurvFrame <- data.frame(time=1:10,status=c(0,1,1,0,0,1,0,0,1,0))
SurvHist <- with(SurvFrame,Hist(time,status))
plot(SurvHist)
plot(SurvHist,box2.col=2,box2.label="experienced\nR user")
plot(SurvHist,
     box2.col=2,
     box1.label="newby",
     box2.label="experienced\nR user",
     oneFitsAll=FALSE,
     arrow1.length=.5,
     arrow1.label="",
     arrow1.lwd=4)

## change the cex of all box labels:
plot(SurvHist,
     box2.col=2,
     box1.label="newby",
     box2.label="experienced\nR user",
     oneFitsAll=FALSE,
     arrow1.length=.5,
     arrow1.label="",
     arrow1.lwd=4,
     label.cex=1)

## change the cex of single box labels:
plot(SurvHist,
     box2.col=2,
     box1.label="newby",
     box2.label="experienced\nR user",
     oneFitsAll=FALSE,
     arrow1.length=.5,
     arrow1.label="",
     arrow1.lwd=4,
     label1.cex=1,
     label2.cex=2)


## The pbc data set from the survival package
library(survival)
data(pbc)
plot(with(pbc,Hist(time,status)),
     stateLabels=c("randomized","transplant","dead"),
     arrowLabelStyle="count")

## two competing risks
comprisk.model <- data.frame(time=1:3,status=1:3)
CRHist <- with(comprisk.model,Hist(time,status,cens.code=2))
plot(CRHist)
plot(CRHist,arrow1.label=paste(expression(eta(s,u))))

plot(CRHist,box2.label="This\nis\nstate 2",arrow1.label=paste(expression(gamma[1](t))))
plot(CRHist,box3.label="Any\nLabel",arrow2.label="any\nlabel")

## change the layout
plot(CRHist,
     box1.label="Alive",
     box2.label="Dead\n cause 1",
     box3.label="Dead\n cause 2",
     arrow1.label=paste(expression(gamma[1](t))),
     arrow2.label=paste(expression(eta[2](t))),
     box1.col=2,
     box2.col=3,
     box3.col=4,
     nrow=2,
     ncol=3,
     box1.row=1,
     box1.column=2,
     box2.row=2,
     box2.column=1,
     box3.row=2,
     box3.column=3)

## more competing risks
comprisk.model2 <- data.frame(time=1:4,status=1:4)
CRHist2 <- with(comprisk.model2,Hist(time,status,cens.code=2))
plot(CRHist2,box1.row=2)

## illness-death models
illness.death.frame <- data.frame(time=1:4,
				  from=c("Disease\nfree",
                                      "Disease\nfree",
                                      "Diseased",
                                      "Disease\nfree"),
				  to=c("0","Diseased","Dead","Dead"))
IDHist <- with(illness.death.frame,Hist(time,event=list(from,to)))
plot(IDHist)

## illness-death with recovery
illness.death.frame2 <- data.frame(time=1:5,
from=c("Disease\nfree","Disease\nfree","Diseased","Diseased","Disease\nfree"),
to=c("0","Diseased","Disease\nfree","Dead","Dead"))
IDHist2 <- with(illness.death.frame2,Hist(time,event=list(from,to)))
plot(IDHist2)

## 4 state models
x=data.frame(from=c(1,2,1,3,4),to=c(2,1,3,4,1),time=1:5)
y=with(x,Hist(time=time,event=list(from=from,to=to)))
plot(y)

## moving the label of some arrows

d <- data.frame(time=1:5,from=c(1,1,1,2,2),to=c(2,3,4,3,4))
h <- with(d,Hist(time,event=list(from,to)))
plot(h,
tagBoxes=TRUE,
stateLabels=c("Remission\nwithout\nGvHD",
    "Remission\nwith\nGvHD",
    "Relapse",
    "Death\nwithout\nrelapse"),
arrowLabelSymbol='alpha',
arrowlabel3.x=35,
arrowlabel3.y=53,
arrowlabel4.y=54,
arrowlabel4.x=68)

##'
prodlim

Product-Limit Estimation for Censored Event History Analysis

v2019.11.13
GPL (>= 2)
Authors
Thomas A. Gerds
Initial release

We don't support your browser anymore

Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.