plot.Lexis {Epi}

Lexis diagrams

Description

The follow-up histories represented by a Lexis object can be plotted using one or two dimensions. The two dimensional plot is a Lexis diagram showing follow-up time simultaneously on two time scales.

Usage

## S3 method for class 'Lexis'
plot(x=Lexis( entry=list(Date=1900,Age=0), exit=list(Age=0) ),
                     time.scale = NULL, type="l", breaks="lightgray", ...)
## S3 method for class 'Lexis'
points(x, time.scale = options()[["Lexis.time.scale"]] , ...)
## S3 method for class 'Lexis'
lines(x, time.scale = options()[["Lexis.time.scale"]], ...)
## S3 method for class 'Lexis'
PY.ann(x, time.scale = options()[["Lexis.time.scale"]], digits=1, ...)

Arguments

x

An object of class Lexis. The default is a bogus Lexis object, so that plot.Lexis can be called without the first argument and still produce a(n empty) Lexis diagram. Unless arguments xlim and ylim are given in this case the diagram is looking pretty daft.

time.scale

A vector of length 1 or 2 giving the time scales to be plotted either by name or numerical order

type

Character indication what to draw: "n" nothing (just set up the diagram), "l" - liefelines, "p" - endpoints of follow-up, "b" - both lifelines and endpoints.

breaks

a string giving the colour of grid lines to be drawn when plotting a split Lexis object. Grid lines can be suppressed by supplying the value NULL to the breaks argument

digits

Numerical. How many digits after the demimal points should be when plotting the person-years.

...

Further graphical parameters to be passed to the plotting methods.

Grids can be drawn (behind the life lines) using the following parameters in plot:

• grid If logical, a background grid is set up using the axis ticks. If a list, the first component is used as positions for the vertical lines and the last as positions for the horizontal. If a nunerical vector, grids on both axes are set up using the distance between the numbers.

• col.grid="lightgray" Color of the background grid.

• lty.grid=2 Line type for the grid.

• coh.grid=FALSE Should a 45 degree grid be plotted?

Details

The plot method for Lexis objects traces “life lines” from the start to the end of follow-up. The points method plots points at the end of the life lines.

If time.scale is of length 1, the life lines are drawn horizontally, with the time scale on the X axis and the id value on the Y axis. If time.scale is of length 2, a Lexis diagram is produced, with diagonal life lines plotted against both time scales simultaneously.

If lex has been split along one of the time axes by a call to splitLexis, then vertical or horizontal grid lines are plotted (on top of the life lines) at the break points.

PY.ann writes the length of each (segment of) life line at the middle of the line. Not advisable to use with large cohorts. Another example is in the example file for occup.

Author(s)

Martyn Plummer

See Also

Lexis, splitLexis

Examples

# A small bogus cohort
xcoh <- structure( list( id = c("A", "B", "C"),
                      birth = c("14/07/1952", "01/04/1957", "10/06/1987"),
                      entry = c("04/08/1965", "08/09/1972", "23/12/1991"),
                       exit = c("27/06/1997", "23/05/1995", "24/07/1998"),
                       fail = c(1, 0, 1) ),
                     .Names = c("id", "birth", "entry", "exit", "fail"),
                  row.names = c("1", "2", "3"),
                      class = "data.frame" )

# Convert the character dates into numerical variables (fractional years)
xcoh$bt <- cal.yr( xcoh$birth, format="%d/%m/%Y" )
xcoh$en <- cal.yr( xcoh$entry, format="%d/%m/%Y" )
xcoh$ex <- cal.yr( xcoh$exit , format="%d/%m/%Y" )

# See how it looks
xcoh

# Define as Lexis object with timescales calendar time and age
Lcoh <- Lexis( entry = list( per=en ),
                exit = list( per=ex, age=ex-bt ),
         exit.status = fail,
                data = xcoh )

# Default plot of follow-up
plot( Lcoh )
# Show follow-up time
PY.ann( Lcoh )

# Show exit status
plot( Lcoh, type="b" )
# Same but failures only
plot( Lcoh, type="b", pch=c(NA,16)[Lcoh$fail+1] )

# With a grid and deaths as endpoints
plot( Lcoh, grid=0:10*10, col="black" )
points( Lcoh, pch=c(NA,16)[Lcoh$lex.Xst+1] )
# With a lot of bells and whistles:
plot( Lcoh, grid=0:20*5, col="black", xaxs="i", yaxs="i",
      xlim=c(1960,2010), ylim=c(0,50), lwd=3, las=1 )
points( Lcoh, pch=c(NA,16)[Lcoh$lex.Xst+1], col="red", cex=1.5 )