analysis-functions: Analysis functions for MCMC outputs

In JGCRI/metrosamp: Basic Metropolis Sampler for Markov Chain Monte Carlo

Description

These functions produce summary statistics, such as expectation values or credible intervals, for MCMC objects.

Usage

MAP(mcrslt, thin = NULL)

EV(mcrslt, thin = NULL)

CI(mcrslt, ci = 0.95, thin = NULL)

rsample(mcrslt, N)

Arguments

mcrslt	A metrosamp object, or a list of metrosamp objects representing multiple chains from the same inference.
thin	Thin mc results to a smaller number of samples before applying the analysis. This can speed up analyses on large sample runs.
ci	The credible interval to produce (default: 95%)
N	Number of samples to draw.

Functions

• MAP: Maximum A-Posteriori parameters

• EV: Expectation value for parameters

• CI: Equal-tailed credibile intervals for parameters

• rsample: Draw random samples from a set of Monte Carlo results

  Samples are drawn from the entire collection of Markov chains, with replacement.

JGCRI/metrosamp documentation built on Aug. 8, 2019, 10:59 p.m.