defaultAlphas: Reshapes alpha or beta vector based on rule and parents

In ralmond/Peanut: Parameterized Bayesian Networks, Abstract Classes

Reshapes alpha or beta vector based on rule and parents

Description

Combination rules can be sorted into multiple-a rules (e.g., Compensatory) and multiple-b rules (e.g., OffsetConjunctive). The function isOffsetRule distinguishes between the two types. These functions adjust the log alpha or beta matrix to the correct length depending on the rule and parents of the node argument.

Usage

PnodeDefaultAlphas(node, rule=PnodeRules(node), link=PnodeLink(node) )
PnodeDefaultBetas(node, rule=PnodeRules(node), link=PnodeLink(node))

Arguments

node	A Pnode object whose PnodeLnAlphas or PnodeBetas field is to be set.
rule	A rule (e.g., Compensatory, OffsetConjunctive or the name of a rule.
link	A link function (e.g., partialCredit) or the name of a link function.

Value

A vector of zeros of a suitable length to be used as a default value for PnodeLnAlphas(node) or PnodeBetas(node).

Note

These are used in the PNetica implementation of the Pnode constructor.

Author(s)

Russell Almond

See Also

Pnode, PnodeLnAlphas, PnodeBetas, isOffsetRule

Examples

## Not run: 
library(PNetica) # Requires PNetica
sess <- NeticaSession()
startSession(sess)


EM1 <- ReadNetworks(system.file("testnets", "PPcompEM.dne",
                                package="PNetica"),
                    session=sess)
EM2 <- ReadNetworks(system.file("testnets", "PPconjEM.dne",
                                package="PNetica"),
                    session=sess)

comp <- PnetFindNode(EM1,"CompensatoryObs")
conj <- PnetFindNode(EM2,"ConjunctiveObs")

stopifnot(
  PnodeDefaultAlphas(comp,"Compensatory") == c(0,0),
  PnodeDefaultBetas(comp,"Compensatory") == 0,
  PnodeDefaultAlphas(conj,"OffsetConjuctive") == 0,
  PnodeDefaultBetas(conj,"OffsetConjunctive") == c(0,0)
)

DeleteNetwork(list(EM1,EM2))
stopSession(sess)

## End(Not run)

ralmond/Peanut documentation built on Sept. 19, 2023, 8:27 a.m.