sampleTriple: Sample the parents of a triple of nodes (Gibbs sampler).

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

Description

Sample from posterior distribution on graph, conditional on all the edges, except for those corresponding to the parents sets of two nodes.

Usage

  sampleTriple(currentNetwork, numberOfNodes, nodesSeq,
    scoresParents, parentsTables, allRows, rowsThatContain,
    logScoreFUN, logScoreParameters, blankBit, bitLengths)

Arguments

currentNetwork	A currentNetwork object
numberOfNodes	The number of nodes in the network. A numeric vector of length 1.
nodesSeq	The vector 1:nNodes(currentNetwork). (Supplied as an argument for possible speed gain)
scoresParents	A list of the form returned by scoreParentsTable()
parentsTables	A list of tables of the form returned by enumerateParentsTable()
allRows	The vector 1:nrow(parentsTables). (Supplied as an argument for possible speed gain)
rowsThatContain	A list of the form created by getRowsThatContain()
logScoreFUN	A list of four elements: offline A function that computes the logScore of a Bayesian Network online A function that incrementally computes the logScore of a Bayesian Network local A function that computes the local logScore of a Bayesian Network prepare A function that prepares the data, and any further pre-computation required by the logScore functions. For Multinomial-Dirichlet models, logScoreMultDirFUN returns the appropriate list; for Normal models with Zellner g-priors, logScoreNormalFUN returns the appropriate list.
logScoreParameters	A list of parameters that are passed to logScoreFUN.

Value

Returns the sampled network. A currentNetwork object.

See Also

BNGibbsSampler, sampleNode

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