mcmc.SCR: Run MCMC algorithm for basic SCR model.

In benaug/SPIM: This package contains MCMC algorithms in R and (some) Rcpp for various Spatial Partial Identity Models. Prospective users should email Ben before using.

Run MCMC algorithm for basic SCR model.

Description

This function runs the MCMC algorithm for the basic SCR model. The data list should have the following elements: 1. y, a n x J capture history 2. X, a matrix with the X and Y trap locations in the first two columns 3. either buff or vertices. buff is the fixed buffer for the traps to produce the state space. It is applied to the minimum and maximum X and Y locations, producing a square or rectangular state space. vertices is a matrix with the X and Y coordinates of a polygonal state space.

Usage

mcmc.SCR(data, niter = 1000, nburn = 0, nthin = 1, M = M,
  inits = inits, proppars = list(lam0 = 0.05, sigma = 0.1, sx = 0.2, sy
  = 0.2), storeLatent = TRUE, Rcpp = TRUE, obstype = "bernoulli")

Arguments

data	a list produced by simSCR or in the same format
niter	number of MCMC iterations to run
nburn	number of MCMC iterations to discard as burn in
nthin	MCMC thinning parameter. Record output on every nthin iterations. nthin=1 corresponds to no thinning
M	The size of the augmented superpopulation
inits	a list of user-supplied initial values. inits=list(psi=psi,lam0=lam0,sigma=sigma)
proppars	a list of tuning parameters for the proposal distributions
storeLatent	a logical indicating whether or not to keep the posteriors for z and s
Rcpp	a logical indicating whether or not to use Rcpp

Value

a list with the posteriors for the SCR parameters (out), s, z

Author(s)

Ben Augustine, Andy Royle

benaug/SPIM documentation built on April 28, 2024, 7:27 a.m.