mcmcapply: Calculations on 'mcmc.list' objects

In datacloning/dclone: Data Cloning and MCMC Tools for Maximum Likelihood Methods

Calculations on 'mcmc.list' objects

Description

Conveniently calculates statistics for mcmc.list objects.

Usage

mcmcapply(x, FUN, ...)
## S3 method for class 'mcmc.list'
stack(x, ...)

Arguments

x	Objects of class mcmc.list.
FUN	A function to be used in the calculations, returning a single value.
...	Other arguments passed to FUN.

Details

mcmcapply returns a certain statistics based on FUN after coercing into a matrix. FUN can be missing, in this case mcmcapply is equivalent to calling as.matrix on an 'mcmc.list' object.

stack can be used to concatenates 'mcmc.list' objects into a single vector along with index variables indicating where each observation originated from (e.g. iteration, variable, chain).

Value

mcmcapply returns statistic value for each variable based on FUN, using all values in all chains of the MCMC object.

stack returns a data frame with columns: iter, variable, chain, value.

Author(s)

Peter Solymos, [e-mail]

Examples

data(regmod)
mcmcapply(regmod, mean)
mcmcapply(regmod, sd)

x <- stack(regmod)
head(x)
summary(x)
library(lattice)
xyplot(value ~ iter | variable, data=x,
    type="l", scales = "free", groups=chain)

datacloning/dclone documentation built on Jan. 7, 2023, 2:38 p.m.