mt_tmcmc_metrop: Simulate a Multiple Try Transformation based Markov Chain...

In kkdey/tmcmcR: tmcmcR : a R/C++ package for Transformation based Markov Chain Monte Carlo

Description

The function simulates a Multiple try TMCMC (MT-TMCMC) chain of length nsamples using the scale, base and burn in taken optimally as default or specified by user. The function outputs the full chain as well as the estimated posterior mean estimated from the samples drawn post burn-in.

Usage

mt_tmcmc_metrop(target_pdf, scale, base, nmove_size, nmove, nsamples,
  revgen = TRUE, burn_in = NULL, verb = FALSE)

Arguments

target_pdf	The log target density function from which the user wants to generate samples.
scale	The proposal density scaling parameter. An approximation of the optimal scaling given the target_pdf is performed by OptimalScaling(). The default scale is this estimated optimal scaling
base	The starting value of the chain
nmove_size	The number of distinct move sizes used.
nmove	The total number of moves proposed.
nsamples	The number of samples to be drawn using the TMCMC algorithm.
revgen	If TRUE, use MT-TMCMC version 1, else version 2. Default is TRUE.
burn_in	The number of samples assigned as burn-in period. The default burn-in is taken to be one-third of nsamples.
verb	logical parameter, if TRUE the function prints the progress of simulation.

Value

Returns a list containing the following items

chain	The full chain produced by the MT-TMCMC method.
post.mean	The estimated posterior mean of the TMCMC chain adjusting for burn-in.

@author Kushal K Dey

@useDynLib tmcmcR @import tmcmcR @export

kkdey/tmcmcR documentation built on May 20, 2019, 10:39 a.m.