initialvalue: Initial values for ECM

In EMMIXSSL: Semi-Supervised Gaussian Mixture Model with a Missing-Data Mechanism

Initial values for ECM

Description

Inittial values for claculating the estimates based on solely on the classified features.

Usage

initialvalue(dat, zm, g, ncov = 2)

Arguments

dat	An n\times p matrix where each row represents an individual observation
zm	An n-dimensional vector containing the class labels including the missing-label denoted as NA.
g	Number of multivariate normal classes.
ncov	Options of structure of sigma matrix; the default value is 2; ncov = 1 for a common covariance matrix; ncov = 2 for the unequal covariance/scale matrices.

Value

pi	A g-dimensional initial vector of the mixing proportions.
mu	A initial p \times g matrix of the location parameters.
sigma	A p\times p covariance matrix if ncov=1, or a list of g covariance matrices with dimension p\times p \times g if ncov=2.

Examples

n<-150
pi<-c(0.25,0.25,0.25,0.25)
sigma<-array(0,dim=c(3,3,4))
sigma[,,1]<-diag(1,3)
sigma[,,2]<-diag(2,3)
sigma[,,3]<-diag(3,3)
sigma[,,4]<-diag(4,3)
mu<-matrix(c(0.2,0.3,0.4,0.2,0.7,0.6,0.1,0.7,1.6,0.2,1.7,0.6),3,4)
dat<-rmix(n=n,pi=pi,mu=mu,sigma=sigma,ncov=2)
xi<-c(-0.5,1)
m<-rlabel(dat=dat$Y,pi=pi,mu=mu,sigma=sigma,xi=xi,ncov=2)
zm<-dat$clust
zm[m==1]<-NA
inits<-initialvalue(g=4,zm=zm,dat=dat$Y,ncov=2)

EMMIXSSL documentation built on Oct. 18, 2022, 5:08 p.m.