TraMineRextras (version 0.2.3)

seqeordplot: Plot the order of event sequences.

Description

Illustrates the order of event sequences in a modified time-series plot. The x axis presents the position in the sequence the y axis the event.

Usage

seqeordplot(seqe,group=NULL,weighted=TRUE,weights=NULL,
            alphabet=NULL,
            type="distinctive",embedding="most-frequent",
            show=c(0,1),hide.col="grey75",
            cpal=NULL,alpha=1,
            lcourse="upwards",lorder="background",
            lweights=TRUE,lwd.min=0.5,lwd.max=4,lty=1,
            cex=1,border=grid.col,border.lwd=grid.lwd/2,
            grid.col="white",grid.fill="grey90",
            grid.scale=1/4,grid.shape="default",grid.lwd=0,
            orderalign=ifelse(split=="last","last","first"),
            split=NULL,layout=NULL,return.data=FALSE,
            main="",sub=NULL,mtext=TRUE,
            xlab="order position",ylab="",
            xlim=NULL,ylim=NULL,...)

Arguments

seqe

an event sequence object as defined by the seqecreate function.

group

A grouping vector.

weighted

Logial. Use of weights in the seqe object.

weights

A weight vector. Overwrites the weights in the seqe object.

alphabet

A vector with event names. Defines the arrangement in the y axis.

type

The trajectories type to draw. Either "distinctive" or "non-embeddable".

embedding

Option for type="non-embeddable", the method how to assign trajectories having multiple corresponding non-embeddable trajectories. Either "most-frequent" (default) or "uniformly".

show

Vector of two values between 0 and 1. Indicate the minimal and maximal relative frequency for a trajectory to be presented in the foreground of the plot.

hide.col

color of trimmed trajectories. If set to "white" trajectories are not shown at all.

cpal

A colour palette to be assigned to the sequences.

alpha

Degree of line and symbol transparency. Choose a number between 0 and 1.

lcourse

Handling for line connection of simultaneous observations. Either "upwards" or "downwards".

lorder

Either "background" (default) or "foreground". The first plots infrequent trajectories, the latter the frequent trajectories in the front.

lweights

logical. Should the line width be proportional to the represented trajectory? If `FALSE`, line width is set as "lwd.min".

lwd.min

The minimal line width to be drawn in the plot.

lwd.max

The maximal line width to be drawn in the plot.

lty

Line type of lines connecting succeeding observations.

cex

Expansion factor for the plotted squared symbols.

border

Color of symbol borders.

border.lwd

Line widths of symbol borders.

grid.col

Color of border of underlaid rectangles.

grid.fill

Color of underlaid rectangles.

grid.scale

Expansion degree of underlaid rectangles.

grid.shape

Either "default" or "proportional".

grid.lwd

Line width or border of underlaid rectangles.

orderalign

Alignment mode for data where the order positions are individual integer orders. align="first" aligns the trajectories left hand, align="last" right hand. Assigning an y category produces an alinement of the first occurrences of this catego

split

Logical value for plot panel arrangement modes. If `TRUE` and `x.orderalign="first"`, the plot produces one plot panel per observed initial event and inscribes the trajectories which are initialised by that event. The case `TRUE` and `align="last

layout

Integer vector of length 2. Determines the number of rows and columns of the plot panels arrangement.

return.data

Returns a summary of the plotted trajectories.

main

A main title for the plot, see also "title".

sub

Subtitles. Used in case of multiple plot panels.

mtext

Logical. Print panel information or not.

xlab

A label for the x axis, defaults to a description of "x".

ylab

A label for the y axis, defaults to a description of "y".

xlim

The x limits (x1, x2) of the plot.

ylim

The y limits (y1, y2) of the plot.

...

Arguments to be passed to methods, such as graphical parameters (see "par").

Examples

## ====================
## plot the biofam data
## ====================

## loading the data and defining an event sequence dataset
## =======================================================

data(biofam)
lab <- c("Parent","Left","Married",
         "Left+Marr","Child","Left+Child",
         "Left+Marr+Child","Divorced")
biofam.seq <- seqdef(data=biofam[,10:25], alphabet=0:7, labels=lab)
## For this example, we consider only a sample of 200 cases
##set.seed(23653)
##sple <- sample(1:nrow(biofam.seq), size=200)
sple <- 500:700  ## need a sample with all elements of the alphabet
##seqstatl(biofam[sple,10:25])
biofam <- biofam[sple,]
biofam.seq <- biofam.seq[sple,]
bf.seqestate <- seqecreate(biofam.seq, tevent = "state")
head(bf.seqestate)

## plot the data
## =============

## distinctive event sequences

par(mar=c(4,8,2,2))
seqeordplot(seqe=bf.seqestate,alphabet=lab)

par(mar=c(4,8,2,2))
seqeordplot(seqe=bf.seqestate,alphabet=lab,
                       lwd.max=6,cex=0.9,show=c(0.05,1))

## non-embeddable sequences

par(mar=c(4,8,2,2))
seqeordplot(seqe=bf.seqestate,alphabet=lab,
                       lwd.max=6,cex=0.9,show=c(0.05,1),
                       type="non-embeddable")

## some additional options

par(mar=c(4,8,2,2)) # how the sequences end
seqeordplot(seqe=bf.seqestate,alphabet=lab,
                       type="non-embeddable",lwd.max=2,
                       orderalign="last",split="last")

par(mar=c(4,8,2,2)) # sequences involving Left+Marr+Child
seqeordplot(seqe=bf.seqestate,alphabet=lab,
                       type="non-embeddable",
                       orderalign="Left+Marr+Child")

par(mar=c(4,8,2,2)) # gender differences
seqeordplot(seqe=bf.seqestate,group=biofam$sex,show=c(0.05,1),
                       alphabet=lab)