bf: Bayes factor between two network structures

In bnlearn: Bayesian Network Structure Learning, Parameter Learning and Inference

Bayes factor between two network structures

Description

Compute the Bayes factor between the structures of two Bayesian networks.

Usage

BF(num, den, data, score, ..., log = TRUE)

Arguments

num, den	two objects of class bn, corresponding to the numerator and the denominator models in the Bayes factor.
data	a data frame containing the data to be used to compute the Bayes factor.
score	a character string, the label of a posterior network score, a BIC scores or "custom-score" for the custom score. If none is specified, the default score is the Bayesian Dirichlet equivalent score ("bde") for discrete networks and the Bayesian Gaussian score ("bge") for Gaussian networks. Other kinds of Bayesian networks are currently supported by using their BIC score to approximate their marginal log-likelihoods.
...	extra tuning arguments for the posterior scores. See score for details.
log	a boolean value. If TRUE, the Bayes factor is given as log(BF).

Value

A single numeric value, the Bayes factor of the two network structures num and den.

Note

The Bayes factor for two network structures, by definition, is the ratio of the respective marginal likelihoods. If we assume the uniform prior over all possible DAGs, it is equivalent to the ratio of the corresponding posterior probabilities. However, note that it is possible to specify different priors using the “...” arguments of BF(). In that case, the value returned by the function will not be the standard Bayes factor.

Author(s)

Marco Scutari

See Also

score, compare, bf.strength.

Examples

data(learning.test)

dag1 = model2network("[A][B][F][C|B][E|B][D|A:B:C]")
dag2 = model2network("[A][C][B|A][D|A][E|D][F|A:C:E]")
BF(dag1, dag2, learning.test, score = "bds", iss = 1)

bnlearn documentation built on July 17, 2026, 5:08 p.m.