MCMCpostsample2D: MCMC posterior sampling for 2D landmark data (Gaussian...

In debashischatterjee111/BPviGM1: Novel Bayesian Procrustes Variance-based Inferences in Geometric Morphometrics

MCMC posterior sampling for 2D landmark data (Gaussian likelihood with Isotropic Error Variance)

Description

Draws posterior from 5 parameters "c1","c2", "b", "theta", "Sigma"

Usage

MCMCpostsample2D(initial, tune, myData, choice, Nsample)

Arguments

initial	The start value of 5*1 parameter vector for MCMC run
tune	Tuning value of MCMC sampler
myData	3D array containing 2 dimensional landmark
choice	Which to compare with 1st object or mean obj
Nsample	Number of MCMC sample desired

Details

"c1","c2" = Location parameter "b" = dilation parameter "theta" =Rotation angle "Sigma" = Isotropic error variation parameter Note that, here we are assuming Isotropy of error variance

Value

matrix containing samples from posterior density of parameter

Examples

## Not run: 
require(shapes)
data(apes)
myData = apes$x
ape10000=MCMCpostsample2D(rnorm(5,1,1),
rep(1,5),apes$x,10,10000)
head(ape10000)

## End(Not run)

debashischatterjee111/BPviGM1 documentation built on April 8, 2023, 7:28 p.m.