sequence (version 2.0)

flux: Draws a flow chart from a transition matrix

Description

Draws a flow chart from a transition matrix on the projection plans of a CA (Correspondence Analysis). requires two objects of class data.frame: tabcoord a table of axis coordinates on up to 7 columns, and tabtr the transition matrix.

Usage

flux(tabcoord, tabtr, dic = NULL, fac = c(1, 2), Sort = NULL, threshold = 0, 
scale = 0.1, cscale = 0.1,main="Transition graph", ...)

Arguments

tabcoord

data.frame: nfac + 3 columns

• id factor: identifier of individuals

• w numeric: weight

• f1 numeric: coordinates on axis 1

• f2,...numeric: next coordinates until nfact

• type factor: "pri" for principal, "ill" for illustrative (or 'supplementary')

tabtr

data.frame: transition matrix. Either frequencies or probabilities. Caution this data.frame must contain the row identifiers as first column

dic

character vector: list of items labels. If NULL (default) must be extractible from tabtr column names

fac

vector: two elements: the numbers of the two factors to use in tabcoord

Sort

numeric: number of the factor used to sort the drawing of the elements. Default = NULL (sorting in the order of the dictionary. Otherwise must be comprised between 1 et nfac. If positive, sort in ascending order. If negative, sort in decreasing order

threshold

numeric: minimal threshold to draw the arrows. Represent a oroportion of the weight of the element used as origin of the arrow.

scale

numeric: Controls the width of the arrows

cscale

numeric: Controls the radius of the circles.

main

character:title of the plot

…

… supplementary arguments for smartArrow

Value

An object of class graph.

Details

tabcoord must be an the attribute $fl or $fc of an object of class ca

References

Pierre, J. S. and C. Kasper (1990). The Design of Ethological Flow-Charts on Factorial Analysis Representations - an Application to the Study of the Male Mole-Cricket Sexual Courtship. Biology of Behaviour 15(3-4): 125-151.(in french) Van der Heijden, P. G. M. 1986. Transition matrices, model fitting and correspondence analysis. In: Data Analysis and Informatics IV (Ed. by E. Diday), pp. 221-226. Elsevier Science Publishers.

See Also

compseq to create transition matrices, compmat to compare transition matrices, read.series.seq to read a file of sequences,ca to perform a Correspondence Analysis

Examples

# NOT RUN {
# Analysis of Wiepkema's data
data(wiepkema)
# Symmetrisation and diagonal loading
symet(wiepkema,ident=TRUE)->swiepkema
# Correspondence Analysis
ca(wiepkema,isup=c(5,11),jsup=c(5,11))->afcwiepk
# Flow chart / transition graph
flux(tabcoord=afcwiepk$fr,tabtr=wiepkema,threshold=0.025,scale=2)

# }