kml (version 2.1.2)

exportClusterization: ~ Function: exportClusterization ~

Description

This function save all the information contain in a Clusterization object.

Usage

exportClusterization(object, y, typeGraph = "bmp", col = "clusters", type = "l", col.mean = "clusters", type.mean = "b", main = "", cex = 1, pch.mean = "letters", pch.time = NA, ...)

Arguments

object

[ClusterizLongData]: object containning the information that have to be saved

y

[couple(numeric)]: indicate which Clusterization shall be exported. The first numeric y[1] is the cluster number, the second y[2] is the rank of clusterization in the list y[1].

typeGraph

[character]: indicate the type of the graph that will be exported. See savePlot for avalaible options

col

[character], [numeric] or vector[numeric]: Specification of the ploting color of the individual trajectories. In addition to the standard possibles values, col="clusters" can be use to color the individual trajectories according to their cl

col.mean

[character], [numeric] or vector[numeric]: Specification of the ploting color of the mean trajectories. In addition to the standard possibles values, col="clusters" can be use to color each mean trajectories according to its clusters.

main

[charater]: give the title of the graph.

type

[character]: what type of plot should be drawn for the individual trajectories ?

type.mean

[character]: what type of plot should be drawn for the mean trajectories ?

cex

[numeric]: fixes the size of the point on the mean trajectories.

pch.mean

[character]: specify the symbol to be used as plotting point on the mean trajectories. Option pch.mean="symbols" or pch.mean="letters" can be used.

pch.time

[vector(numeric)]: precise the time at which a point should be plot (usefull if there is a important number of time, see plot for detail).

...

Graphical parameters to be passed to methods, see plot(LongData) and par for details.

Value

• Two files and two graphes.

Details

exportClusterizationis meanly used by the function choice. It export the clusters affectation (individual and its letters) in a file named "objectName-Clusters.csv", the information about the clusterization (algortihm used, clusters' number, convergence time, percent in each cluster, criterion quality name and value, imputation method and starting condition) in another file names "objectName-Details.csv". It also save two graphs, one with the trajectories, the other one with the subgroups.

Examples

#############
### Creating a ClusterizLongData object, with 3 cluterization (5 clusters each)
dn <- as.cld(gald())
kml(dn,5,3)

### Exporting the second clusterization in pdf format
try(exportClusterization(dn,c(5,2),typeGraph="pdf"))