Pancho Aguirre
2010-Apr-05 23:22 UTC
[R] Agnes in Cluster Package and index.G1 in the clusterSim package questions
Dear R Users:
I am new to R and I am trying to do a cluster analysis on a
single continuous variable using the Agnes [Agglomerative Nesting (Hierarchical
Clustering) ] in the Package ‘cluster’. I was able to apply this clustering
method to
my data:
ward1 <- Agnes(balances, diss= FALSE, metric "euclidean", stand =
TRUE, method = "ward", keep.diss
=TRUE, keep.data = TRUE)
Now I am wondering if someone can point me in the right direction
with the following:
1. 1. How do I determine the number of clusters? I would like to apply
the Calinkski and
Harabasz (1974) stopping rule recommended in the Monte Carlo Simulation by
Milligan
and Cooper (1985).
2. 2. How do I obtain the clustering rules? (I.e. how
do I assign my observations to the cluster from my final solution?).
I was able to find the Package ‘clusterSim’ which Calculates
Calinski-Harabasz pseudo F-statistic but I am having some difficulty
substituting in the CL (A vector of integers indicating the cluster to which
each object is allocated) argument in
the formula:
index.G1 (x,cl,d=NULL,centrotypes="centroids")
Thanks in advance for any help,
Sincerely,
Pancho Aguirre
[[alternative HTML version deleted]]
