postclassif {Geneclust}R Documentation

Posterior membership probabilities and cluster assignment

Description

Post-processing of MCMC outputs computing individual membership coefficients and the most likely cluster assignment.

Usage

postclassif(path.mcmc, coordinates, popmbrship=NULL, plot=TRUE, print=FALSE, file=path.mcmc, write=FALSE)

Arguments

path.mcmc Path to the MCMC program output files directory
coordinates Spatial coordinates of individuals. A matrix with 2 columns and one line per individual
popmbrship A vector with number of individuals components containing the true population memberships (if they are known!). If popmbrship=NULL, the population of origin of each individuals is unknown.
plot Logical: if TRUE, maps are plotted
print Logical: if TRUE, maps are also printed.
file Character : Path to file where maps should be printed
write Logical: If TRUE, data are written in ascci files named "postmode.indiv.txt" and "proplab.txt" respectively

Value

A list with components:

postmode.indiv A vector that contains the cluster assignment of each individual
effpop.est A vector which contains the number of individuals in each population defined
proplab Matrix of posterior membership probabilities for each individuals
K.est The number of populations defined from the cluster assignment of each individual
errclassif A rate of misclassification computed if the true population memberships are given as argument
K.distrib The posterior distribution of K (number of populations)
modeK The mode of the posterior distribution of K (number of populations)

Author(s)

Sophie Ancelet

Examples


# library(Geneclust)

## see the example described in the function mcmcgeneclust

[Package Geneclust version 1.0.0 Index]