logScoreMultDir.bn: Compute the log marginal likelihood of the supplied Bayesian...

In rjbgoudie/structmcmc: Structural inference for Bayesian Networks and Variable Selection.

Description

The data must be discrete. The conditional distributions of each random variable, conditional on its parents are assumed to be multinomial, with Dirichlet priors for the parameters.

Usage

  ## S3 method for class 'bn'
 logScoreMultDir(x, data,
    cache = new.env(hash = T), hyperparameters = "bdeu",
    checkInput = T, ...)

Arguments

x	An object of class "bn". The Bayesian Network by for which the marginal likelihood is computed.
data	A data.frame, with columns being factors giving the values of each random variable.
cache	Optionally, provide an environment with cached local scores for this data.
hyperparameters	A character vector of length one. Either "bdeu", "qi", "one", or "point9"
checkInput	A logical of length 1, specifying whether to check the inputs to the function.
...	Further arguments, currently unused

Details

The notation here roughly follows Mukherjee and Speed (2008) Network inference using informative priors. PNAS 105 (38) 14313-14318, doi: 10.1073/pnas.0802272105

Value

A numeric vector of length 1, giving the log marginal likelihood. The environment 'cache' will also be updated because its scope is global.

See Also

logScoreMultDir, logScoreMultDir.bn.list

Examples

x1 <- factor(c(1, 1, 0, 1, 0, 0, 1, 0, 1, 0))
x2 <- factor(c(0, 1, 0, 1, 0, 1, 1, 0, 1, 0))
x3 <- factor(c(0, 1, 1, 1, 0, 1, 1, 0, 1, 0))
data <- data.frame(x1 = x1, x2 = x2,  x3 = x3)
logScoreMultDir(bn(c(), c(1), c(2)), data)

rjbgoudie/structmcmc documentation built on Nov. 3, 2020, 3:41 a.m.