DPM.HODC: Hierachical ordered density clustering (HODC) Algorithm with...

In BANFF: Bayesian Network Feature Finder

Description

Hierachical ordered density clustering (HODC) Algorithm with input generated by DPdensity

Usage

DPM.HODC(v, pvalue, DPM.mcmc = list(nburn = 2000, nsave = 1, nskip = 0,
  ndisplay = 10), DPM.prior = list(a0 = 2, b0 = 1, m2 = rep(0, 1), s2 =
  diag(1e+05, 1), psiinv2 = solve(diag(0.5, 1)), nu1 = 4, nu2 = 4, tau1 = 1,
  tau2 = 100))

Arguments

v	number of iterations set for DPM fitting by "DPdensity"
pvalue	a vector of p-values obtained from large scale statistical hypothesis testing
DPM.mcmc	list
DPM.prior	list

Details

Without the information of networking, we can have an approximation to the marginal density by DPM model fitting on r. Suppose the number of finite mixture normals is equal to L_0+L_1, which means the number of classes we have, we apply HODC algorithm in partitioning the $L_0$ and $L_1$ components into two classes. For this function, the input is generated by Mclust

Value

a list of HODC algorithm returned parameters.

mean

the means of each of two cluster for every DPM fitting by "DPdensity"

mu0

the means of the cluster with smaller mean

mu1

the means of the cluster with larger mean

variance

the variance of each of two cluster for every DPM fitting by "DPdensity"

var0

the variances of the cluster with smaller mean

var1

the variances of the cluster with larger mean

probability

the probability of each of two cluster for every DPM fitting by "DPdensity"

pro0

the probabilities of the cluster with smaller mean

pro1

the probabilities of the cluster with larger mean

classification

The classification corresponding to each cluster for every DPM fitting by "DPdensity"

Examples

## Not run: 
###random make the density
rstat=c(rnorm(50,mean=1),rnorm(50,mean=2),rnorm(100,mean=4),rnorm(100,mean=8))
###transformed into pvalue
pvalue=pnorm(-rstat)
dpdensityHODC=DPM.HODC(v=5,pvalue)

## End(Not run)

BANFF documentation built on May 29, 2017, 11:59 a.m.