MFclust {MFDA}R Documentation

Model-Based Functional Data Clustering

Description

Clustering via rejection-controlled EM initialized by kmeans clustering for functional Gaussian mixture models. The number of clusters and the clustering model is chosen to minimize the BIC.

Usage

MFclust(data, minG, maxG, nchain=NULL,thres=NULL,iter.max=NULL,my.alpha=NULL, ...)

Arguments

data A numeric vector, matrix, or data frame of observations. Categorical variables are not allowed. If a matrix or data frame, rows correspond to observations and columns correspond to time.
minG An integer vector specifying the minimum number of mixture components (clusters) to be considered. The default is 1 component.
maxG An integer vector specifying the maximum number of mixture components (clusters) to be considered. The default is 9 components.
nchain An integer specifying the number of Markov chains in RCEM. The default is 5 chains.
thres A number between 0 and 1 specifying the threshold value of rejection step in RCEM. The default is 0.5.
iter.max An iteger specifying the maximum number of iteration in RCEM. The default is 10.
my.alpha The prior for the cluster proportion. The default is 1.
... The arguments to be part of the function.

Value

A list representing the best model (according to BIC) for the given range of numbers of clusters. The following components are included:

BIC A vector giving the BIC value for each number of clusters.
nclust A scalar giving the optimal number of clusters.
clust A vector whose ith entry is the indicator that observation i belongs to the k component in the model.
clust.center A matrix whose rows are the means of each group.

References

Ma, P., Castillo-Davis, C., Zhong, W., and Liu, J. S. (2006) A data-driven clustering method for time course gene expression data, Nucleic Acids Research, 34 (4), 1261-1269.

See Also

MFclust.compute

Examples

data("testdata")
my.clust<-MFclust(testdata,minG=3,maxG=5,nchain=1,iter.max=1)


[Package MFDA version 1.1-1 Index]