UpperBoundLogLikelihood: Upper bound of the loglikelihood

In MoTBFs: Learning Hybrid Bayesian Networks using Mixtures of Truncated Basis Functions

Upper bound of the loglikelihood

Description

Computes an upper bound of the expected loglikelihood of a dataset given a randomly generated MoTBF density.

Usage

UpperBoundLogLikelihood(f, data, min, max)

Arguments

f	A function to evaluate of class "character", "motbf" or others.
data	A "numeric" array which contains the values to evaluate.
min	A "numeric" value giving the lower limit of the domain.
max	A "numeric" value giving the upper limit of the domain.

Value

A "numeric" value which is the log-likelihood of the evaluated ramdom function.

See Also

getNonNormalisedRandomMoTBF

Examples

data <- rnorm(20)
f <- getNonNormalisedRandomMoTBF(degree = 8, POTENTIAL_TYPE = "MOP")
UpperBoundLogLikelihood(f, data, min = -2.5, max = 3.2)

data <- rexp(20)
f <- getNonNormalisedRandomMoTBF(degree = 8, POTENTIAL_TYPE = "MTE")
UpperBoundLogLikelihood(f, data, min = 0, max = 5)

MoTBFs documentation built on April 18, 2022, 5:06 p.m.