EMGrank: EMGrank

In valse: Variable Selection with Mixture of Models

Description

Run an generalized EM algorithm developped for mixture of Gaussian regression models with variable selection by an extension of the low rank estimator. Reparametrization is done to ensure invariance by homothetic transformation. It returns a collection of models, varying the number of clusters and the rank of the regression mean.

Usage

EMGrank(Pi, Rho, mini, maxi, X, Y, eps, rank, fast)

Arguments

Pi	An initialization for pi
Rho	An initialization for rho, the variance parameter
mini	integer, minimum number of iterations in the EM algorithm, by default = 10
maxi	integer, maximum number of iterations in the EM algorithm, by default = 100
X	matrix of covariates (of size n*p)
Y	matrix of responses (of size n*m)
eps	real, threshold to say the EM algorithm converges, by default = 1e-4
rank	vector of possible ranks
fast	boolean to enable or not the C function call

Value

A list (corresponding to the model collection) defined by (phi,LLF): phi : regression mean for each cluster, an array of size p*m*k LLF : log likelihood with respect to the training set

valse documentation built on May 31, 2021, 9:10 a.m.