buildDescriptive: Build a discrete Bayesian descriptive network.

In MNLR/BNWeatherGen: Weather Generators with Bayesian Networks

Build a discrete Bayesian descriptive network.

Description

A descriptive Bayesian network that approximates the Joint Probability of the dataset y, withoyt taking into account the temporal aspect.

Usage

buildDescriptive(
  y,
  structure.learning.algorithm = "tabu",
  structure.learning.args.list = list(),
  compile.junction = FALSE,
  param.learning.method = "bayes",
  parallelize = FALSE,
  cluster.type = "FORK",
  n.cores = NULL
)

Arguments

y	Stations dataset, as output by loadeR::loadStationData()
structure.learning.algorithm	Algorithm used to perform structure learning, with name as text. Supports all the score-based, constraint-based and hybrid bayesian network structure learning algorithms from bnlearn. Refer to Details for a list of supported algorithms.
structure.learning.args.list	List of arguments passed to structure.learning.algorithm, in particular distance argument if local learning is used. Note that other arguments, e.g. whitelist, are an option (check the naming convention, see Details). Refer to bnlearn for the specific options.
compile.junction	Compile the junction tree from BN.fit to compute probabilities. Can be set to FALSE. Compiling the junction tree is necessary for using exact inference at the simulating stage.
param.learning.method	Either "bayes", for bayesian estimation; or "mle", for Maximum Likelihood Estimation.
parallelize	Set to TRUE for parallelization. Refer to the parallel and see Details.
cluster.type	Either "PSOCK" or "FORK". Use the former under Windows systems, refer to parallel package.
n.cores	When parallelize = TRUE, number of threads to be used, will use detectCores()-1 if not set.

Details

buildDescriptive() can be used to build a Bayesian network that characterizes the spatial aspects of the dataset, whithout taking into account autocorrelation of the weather series.

Author(s)

Mikel N. Legasa

MNLR/BNWeatherGen documentation built on June 2, 2023, 9:02 p.m.