summary.postpr: Posterior model probabilities and Bayes factors

In bcm-uga/abc: Tools for Approximate Bayesian Computation (ABC)

Description

This function extracts the posterior model probabilities and calculates the Bayes factors from an object of class "postpr".

Usage

## S3 method for class 'postpr'
summary(object, rejection = TRUE, print = TRUE, digits
= max(3, getOption("digits")-3), ...)

Arguments

object	an object of class "postpr".
rejection	logical, if method is "mnlogistic" or "neuralnet", should the approximate model probabilities based on the rejection method returned.
print	logical, if TRUE prints the mean models probabilities.
digits	the digits to be rounded to.
...	other arguments.

Value

A list with the following components if method="rejection":

Prob	an object of class table of the posterior model probabilities.
BayesF	an object of class table with the Bayes factors between pairs of models.

A list with the following components if method is "mnlogistic" or "neuralnet" and rejection is TRUE:

rejection	a list with the same components as above
mnlogistic	a list with the same components as above

See Also

postpr

Examples

## see ?postpr for examples

bcm-uga/abc documentation built on May 11, 2019, 3:03 p.m.