em.mv.gmm: E-M Algorithm for Multivariate Gaussian Mixture Model // HL

In nwakim/nwREM: Implementation of an L0 penalized mvGMM EM algorithm

Description

E-M Algorithm for Multivariate Gaussian Mixture Model // HL

Usage

em.mv.gmm(x, k, max.iter = 10000, tol = 1e-08)

Arguments

x	A (n x d) matrix of observed data // HL
k	The number of mixing components // HL
max.iter	Maximum number of iterations
tol	Relative tolerance of likelihood at convergence.

Value

A list containing the following values: * llk : log-likelihood value * lambdas : size k vector of estimated mixing proportions * mus : (d x k) of estimated means * covs : (d x d x k) vectors of estimated standard deviations * iter : number of iterations

nwakim/nwREM documentation built on May 22, 2019, 5:34 p.m.