output.mcmc {BPEC}R Documentation

Load location data

Description

Provides various MCMC tuning parameters, as well as posterior samples for convergence assessment.

Usage

output.mcmc(bpecout)

## S3 method for class 'bpec'
output.mcmc(bpecout)

Arguments

bpecout

R object from bpec.mcmc run

Value

MCMCparams

Various MCMC tuning parameters, useful for development.

codaInput

Posterior samples from the two MCMC chains for the cluster means, cluster covariance entries, as well as the root haplotype. Note that, since the number of clusters varies from iteration to iteration, some samples are simply draws from the prior (corresponding to empty clusters). This variable can be loaded directly into the coda package for convergence analysis.

Author(s)

Ioanna Manolopoulou & Axel Hille

Examples

## if you want to load the `mini' example Brown Frog dataset
data(MacrocnemisRawSeqs)
data(MacrocnemisCoordsLocsMini)
rawSeqs <- MacrocnemisRawSeqs
coordsLocs <- MacrocnemisCoordsLocsMini

dims <- 3 #this is 2 if you only have geographical longitude/latitude. 
#(add 1 for each environmental or phenotypic covariate)
maxMig <- 2 #you will need a higher maximum number of migrations, suggest 7
ds <- 0 #start with ds=0 and increase to 1 and then to 2
iter <- 1000 #you will need far more iterations for convergence, start with 100,000
postSamples <- 100 #you will need at least 100 saved posterior samples

#run the Markov chain Monte Carlo sampler
bpecout <- bpec.mcmc(rawSeqs,coordsLocs,maxMig,iter,ds,postSamples,dims)
output.mcmc(bpecout)


[Package BPEC version 1.3.1 Index]