mixalg.EM: EM algorithm

In CAMAN: Finite Mixture Models and Meta-Analysis Tools - Based on C.A.MAN

EM algorithm

Description

In the fixed support size case the number of components k is assumed to be known. Here the unknown parameters are the mixing weights p[j] and the parameters lambda[j] of the subpopulation.

Estimation of these models' parameters is usually achieved by application of the EM algorithm.

Usage

mixalg.EM (mix = NULL, p, t, obs = NULL, weights = NULL, family = "gaussian", 
    data = NULL, pop.at.risk = NULL, var.lnOR = NULL, limit = 0.01, 
    acc = 10^(-7), numiter = 5000)    

Arguments

mix	A CAMAN-object which quantifies a finite mixture model and the input data.
p	vector containing the starting values for p
t	vector containing the starting values for the distribution specific parameters (poisson->\lambda, gaussian-> \mu)
obs	observed / dependent variable. Vector or colname of data. Must be specified!
weights	weights of the data. Vector or colname of data. Default is NULL.
family	the underlying type density function as a character ("gaussian", "poisson" or "binomial")!
data	an optional data frame. obs, weights, pop.at.risk and var.lnOR can be specified as column name of the data frame.
pop.at.risk	population at risk: These data could be used to determine a mixture model for Poisson data. Vector or colname of data. Default isNULL.
var.lnOR	variances of the data: These variances might be given when working with meta analyses! Vector or colname of data. Default is NULL.
limit	parameter to control the limit of union several components. Default is 0.01.
acc	convergence criterion. VEM and EM loops stop when deltaLL<acc. Default is 10^(-7).
numiter	parameter to control the maximal number of iterations in the VEM and EM loops. Default is 5000.

Value

The function returns a CAMAN.object object, thus the same as mixalg!

Author(s)

Peter Schlattmann and Johannes Hoehne

Examples

data(vitA)
m1<-mixalg.EM(obs="logrr",var.lnOR="var" ,family="gaussian",
              p=c(1),t=c(0),data=vitA)
m2<-mixalg.EM(obs="logrr",var.lnOR="var" ,family="gaussian",
              p=c(0.5,0.5),t=c(-0.3,0.2),data=vitA)

# apply EM-algorithm on an existing CAMAN.object: 
data(thai_cohort)
mix0 <- mixalg(obs="counts", weights="frequency", family="poisson", 
               data=thai_cohort, numiter=18000, acc=0.00001,startk=25)
em0<-mixalg.EM(mix0,p=c(1),t=c(1))
em1<-mixalg.EM(mix0,p=c(0.7,0.3),t=c(2,9))

CAMAN documentation built on Sept. 22, 2023, 5:12 p.m.