plot.metropolis: A Function to Plot metropolis objects

In FAmle: Maximum Likelihood and Bayesian Estimation of Univariate Probability Distributions

A Function to Plot metropolis objects

Description

This function allows to user to call different plots for visual assessment of the posterior distribution(s).

Usage

## S3 method for class 'metropolis'
plot(x, plot.type = "carlin", pos = 1:x$iter, ...)

Arguments

x	mle object corresponding to the fitted model.
plot.type	The user may choose betweew: carlin returns the same plot as in Carlin and Louis (2009) (see References); ts returns plot.ts; pairs returns a pairs; hist returns an hist for each marginal posterior distribution; post.pred returns an histogram of the data's posterior predictive distribution.
pos	May be used by the user to plot a subset (i.e. a random subset, sample)) of the posterior distribution when pairs is called. This avoids using too much memory while building the plot.
...	Additional arguments pertaining to function plot.default.

References

See list of references for metropolis.

See Also

metropolis

Examples

data(yarns)
x <- yarns$x
fit.x <- mle(x,'gamma',c(.1,.1))
bayes.x <- metropolis(model=fit.x,iter=100,
	trans.list=list(function(x) exp(x),function(x) exp(x)))
plot(bayes.x)
plot(bayes.x,'hist',col='cyan')
plot(bayes.x,'pairs',cex=.1,pch=19)
plot(bayes.x,'pairs',pos=sample(1:bayes.x$iter,20),col='red')
plot(bayes.x,'post.pred',col='green')

FAmle documentation built on March 18, 2022, 5:29 p.m.