CODEWITHSUNDEEP
rsequencecluster-analysistraminer
jbunk
jbunk

Reputation: 23

Displaying Frequency Table of Individual Clusters Using TraMiner

I am currently doing a clustering exercise using TraMiner. I have a frequency table, df.seq, that I broke into four clusters using the following code:

library(cluster)
df.om <- seqdist(df.seq, method='OM', indel=1, sm='TRATE', with.missing=TRUE)
clusterward <- agnes(df.om, diss=TRUE, method="ward")
df.cl4 <- cutree(clusterward, k=4)
cl4.lab <- factor(df.cl4, labels=paste("Cluster", 1:4))

Then, I plot a sequence frequency plot of my four clusters. 

seqfplot(df.seq, group=cl4.lab, pbarw=FALSE, border=NA, withlegend = FALSE, yaxis = "pct", cpal=palette(rainbow(length(unique(df.new$cart)))))

While the seqfplot gives a good visual, I would like to see a frequency table for each individual cluster. For example, I can do:

seqtab(df.seq)

and get the following output:

Sequence Frequency %

Item #1 10 30%

Item #2 9 25%

Item #3 8 20%

Any help would be extremely appreciated!

Upvotes: 1

Views: 763

Answers (1)

Gilbert
Gilbert

Reputation: 3669

You can use by. I illustrate using the actcal data that ships with TraMineR

library(TraMineR)
data(actcal)
actcal.seq <- seqdef(actcal[,13:24])
seqfplot(actcal.seq, group=actcal$sex)
by(actcal.seq, actcal$sex, seqtab)

For your example the command would read

by(df.seq, cl4.lab, seqtab)

Hope this helps.

Upvotes: 1

Related Questions