MWG: Metropolis within Gibbs algorithm for srplv

In agathasr/srplv: Parametric models for estimating the lifetime distribution of a component from superposed renewal processes by means of latent variables.

Description

MWG obtains MCMC sample from posterior distribution via MWG algorithm for superposed renewal processes by means of latent variables.

Usage

MWG(guess, Data, distribution, burn, jump, prior1, prior2, n.size = 1000)

Arguments

guess	The vector of initial values for the parameters. The default is (1,1).
Data	Data in the srplv format as data_to_srplv.
distribution	The chose distribution: weibull, gamma, lnorm, llogis.
burn	Burn-in sample.
jump	Jump sample.
prior1	Hyperparameters for the prior distribution for parameter 1.
prior2	Hyperparameters for the prior distribution for parameter 2.
n.size	Number of MCMC sample. The default is 1000.

Value

A list with the following components:

post.values	The sample from the posterior distribution for each parameter (column) with size n.size.
acceptance	The acceptance rate.
time	The computational time.
matrix.cpo	Matrix with dimension n.sys by n.size containing Conditional Predictive Ordinate (CPO) values for n.sys systems and n.size posterior sample.

agathasr/srplv documentation built on May 17, 2020, 12:21 a.m.