em.clust {MFDA}R Documentation

Model-Based Clustering for a single Markov chain

Description

EM algorithm for functional Gaussian mixture models.

Usage

em.clust(x,clust,mu,zeta,varht,thres,iter.max,alpha)

Arguments

x A numeric matrix, or data frame of observations, rows correspond to functional observations and columns correspond to the number of repeated measurements.
clust An integer vector specifying the initial clustering membership.
mu A matrix specifying the initial estimate of cluster mean profile.
zeta A numerical vector specifying the initial estimate of cluster precision parameters.
varht A numerical vector specifying the initial estimate of cluster error variance.
thres A threshold value for rejection-controlled step.
iter.max An integer limit on the number of EM iterations.
alpha The prior for the cluster proportions p.

Value

A list contains estimated cluster membership, estimated cluster mean profile, Bayesian Information Criterion, negative loglikelihood.

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.


[Package MFDA version 1.1-1 Index]