CAMAN (version 0.78)

vem_grad : VEM algorithm for univariate data, for bivariate data and for meta data

Description

This function implements a VEM algorithm.

Usage

vem_grad(obs1, obs2, type, data, var1, var2, 
         lambda1, lambda2, p, startk, 
         numiter=5000, acc=1.e-7)

Arguments

obs1

the first column of the observations

obs2

the second column of the observations

type

kind of data

data

an optional data frame

lambda1

Means of the first column of the observations

lambda2

Means of the second column of the observations

p

Mixing weight

var1

Variance of the first column of the observations(only for meta-analysis)

var2

Variance of the second column of the observations (only for meta-analysis)

startk

starting/maximal number of components. This number will be used to compute the grid in the VEM. Default is 20.)

numiter

parameter to control the maximal number of iterations in the VEM and EM loops. Default is 5000.

acc

convergence criterion. Default is 1.e-7

Examples

if (FALSE) {
#1.	VEM-algorithm  for bivariate normally distributed data 
#Examples
data(rs12363681)
vem_grad(obs1=x,obs2=y,type="bi", data=rs12363681,startk=20)
#2.VEM for metadata
data(CT)
vem_grad(obs1=logitTPR, obs2= logitTNR, 
         var1= varlogitTPR, var2= varlogitTNR,
         type="meta", data=CT, startk=20)
}