scoreagainstDAG {BiDAG}

Calculating the score of a sample against a DAG

Description

This function calculates the score of a given sample against a DAG represented by its incidence matrix.

Usage

scoreagainstDAG(
  scorepar,
  incidence,
  datatoscore = NULL,
  marginalise = FALSE,
  onlymain = FALSE,
  bdecatCvec = NULL
)

Arguments

scorepar

an object of class scoreparameters; see constructor function scoreparameters

incidence

a square matrix of dimensions equal to the number of variables with entries in {0,1}, representing the adjacency matrix of the DAG against which the score is calculated

datatoscore

(optional) a matrix (vector) containing binary (for BDe score) or continuous (for the BGe score) observations (or just one observation) to be scored; the number of columns should be equal to the number of variables in the Bayesian network, the number of rows should be equal to the number of observations; by default all data from scorepar parameter is used

marginalise

(optional for continuous data) defines, whether to use the posterior mean for scoring (default) or to marginalise over the posterior distribution (more computationally costly)

onlymain

(optional), defines the the score is computed for nodes excluding 'bgnodes'; FALSE by default

bdecatCvec

(optional for categorical data)

Value

the log of the BDe/BGe score of given observations against a DAG

Author(s)

Jack Kuipers, Polina Suter

References

Heckerman D and Geiger D, (1995). Learning Bayesian networks: A unification for discrete and Gaussian domains. In Eleventh Conference on Uncertainty in Artificial Intelligence, pages 274-284, 1995.

Examples

 Asiascore<-scoreparameters("bde", Asia[1:100,]) #we wish to score only first 100 observations
 scoreagainstDAG(Asiascore, Asiamat)