readMCMC: MCMC Reader

In BaM-tools/RBaM: Bayesian Modeling: Estimate a Computer Model and Make Uncertain Predictions

MCMC Reader

Description

Read raw MCMC samples, return cooked (burnt & sliced) ones

Usage

readMCMC(
  file = "Results_Cooking.txt",
  burnFactor = 0,
  slimFactor = 1,
  sep = "",
  reportFile = NULL,
  panelPerCol = 10,
  panelHeight = 3,
  panelWidth = 23/panelPerCol
)

Arguments

file	Character, full path to MCMC file.
burnFactor	Numeric, burn factor. 0.1 means the first 10 are discarded.
slimFactor	Integer, slim factor. 10 means that only one iteration every 10 is kept.
sep	Character, separator used in MCMC file.
reportFile	Character, full path to pdf report file, not created if NULL
panelPerCol	Integer, max number of panels per column
panelHeight	Numeric, height of each panel
panelWidth	Numeric, width of each panel

Value

A data frame containing the cooked mcmc samples.

Examples

# Create Monte Carlo samples and write them to file
n=4000
sim=data.frame(p1=rnorm(n),p2=rlnorm(n),p3=runif(n))
workspace=tempdir()
write.table(sim,file=file.path(workspace,'MCMC.txt'),row.names=FALSE)
# Read file, burn the first half and keep every other row
M=readMCMC(file=file.path(workspace,'MCMC.txt'),burnFactor=0.5,slimFactor=2)
dim(M)

BaM-tools/RBaM documentation built on April 11, 2025, 10:01 p.m.