index.G1 {clusterSim}

Calculates Calinski-Harabasz pseudo F-statistic

Description

Calculates Calinski-Harabasz pseudo F-statistic

Usage

index.G1 (x,cl,d=NULL,centrotypes="centroids")

Arguments

x

data

cl

A vector of integers indicating the cluster to which each object is allocated

d

optional distance matrix, used for calculations if centrotypes="medoids"

centrotypes

"centroids" or "medoids"

Details

See file ../doc/indexG1_details.pdf for further details.

thank to Nejc Ilc from University of Ljubljana for fixing error for one-element clusters.

Value

Calinski-Harabasz pseudo F-statistic

Author(s)

Marek Walesiak [e-mail], Andrzej Dudek [e-mail]

Department of Econometrics and Computer Science, University of Economics, Wroclaw, Poland http://keii.ue.wroc.pl/clusterSim/

References

Calinski, T., Harabasz, J. (1974), A dendrite method for cluster analysis, "Communications in Statistics", vol. 3, 1-27. Available at: doi:10.1080/03610927408827101.

Everitt, B.S., Landau, E., Leese, M. (2001), Cluster analysis, Arnold, London, p. 103. ISBN 9780340761199.

Gatnar, E., Walesiak, M. (Eds.) (2004), Metody statystycznej analizy wielowymiarowej w badaniach marketingowych [Multivariate statistical analysis methods in marketing research], Wydawnictwo AE, Wroclaw, p. 338. Available at: http://keii.ue.wroc.pl/pracownicy/mw/2004_Gatnar_Walesiak_Metody_SAW_w_badaniach_marketingowych.pdf.

Gordon, A.D. (1999), Classification, Chapman & Hall/CRC, London, p. 62. ISBN 9781584880134.

Milligan, G.W., Cooper, M.C. (1985), An examination of procedures of determining the number of cluster in a data set, "Psychometrika", vol. 50, no. 2, 159-179. Available at: doi:10.1007/BF02294245.

See Also

index.G2,index.G3,index.S, index.C, index.H,index.KL,index.Gap, index.DB

Examples

# Example 1
library(clusterSim)
data(data_ratio)
c<- pam(data_ratio,10)
index.G1(data_ratio,c$clustering)

# Example 2
library(clusterSim)
data(data_ratio)
md <- dist(data_ratio, method="euclidean")
# nc - number_of_clusters
min_nc=2
max_nc=20
res <- array(0,c(max_nc-min_nc+1,2))
res[,1] <- min_nc:max_nc
clusters <- NULL
for (nc in min_nc:max_nc)
{
cl2 <- pam(md, nc, diss=TRUE)
res[nc-min_nc+1,2] <- G1 <- index.G1(data_ratio,cl2$cluster,centrotypes="centroids")
clusters <- rbind(clusters, cl2$cluster)
}
print(paste("max G1 for",(min_nc:max_nc)[which.max(res[,2])],"clusters=",max(res[,2])))
print("clustering for max G1")
print(clusters[which.max(res[,2]),])
#write.table(res,file="G1_res.csv",sep=";",dec=",",row.names=TRUE,col.names=FALSE)
plot(res, type="p", pch=0, xlab="Number of clusters", ylab="G1", xaxt="n")
axis(1, c(min_nc:max_nc))