conditionalmotbf.learning: Learning Conditional Functions

Description

Collection of functions for learning conditional MoTBFs, computing the internal BIC, selecting the parents that get the best BIC value, and other internal functions needed to get the final conditionals.

Usage

conditionalMethod(
  data,
  nameParents,
  nameChild,
  numIntervals,
  POTENTIAL_TYPE,
  maxParam = NULL,
  s = NULL,
  priorData = NULL
)
conditional(
  data,
  nameParents,
  nameChild,
  domainChild,
  domainParents,
  numIntervals,
  mm,
  POTENTIAL_TYPE,
  maxParam = NULL,
  s = NULL,
  priorData = NULL
)
select(
  data,
  nameParents,
  nameChild,
  domainChild,
  domainParents,
  numIntervals,
  POTENTIAL_TYPE,
  maxParam = NULL,
  s = NULL,
  priorData = NULL
)
learn.tree.Intervals(
  data,
  nameParents,
  nameChild,
  domainParents,
  domainChild,
  numIntervals,
  POTENTIAL_TYPE,
  maxParam = NULL,
  s = NULL,
  priorData = NULL
)
BICscoreMoTBF(conditionalfunction, data, nameParents, nameChild)

Arguments

data

A dataset of class "data.frame".

nameParents

A "character" vector with the names of the parent variables.

nameChild

The name of the child variable as a "character" string.

numIntervals

A "numeric" value indicating the maximum number of intervals in which we want to split the domain of the parent variables.

POTENTIAL_TYPE

A "character" string specifying the posibles potential types, must be one of "MOP" or "MTE". You can specify just the initial letter.

maxParam

A "numeric" value which indicate the maximum number of coefficients in the function. By default it is NULL; if not, the output is the function which gets the best BIC with at most this number of parameters.

A "numeric" coefficient which fixes the confidence of the prior knowledge we are going to introduce. By default it is NULL, only we must modify it if we want to incorporate prior information to the fits.

priorData

Prior dataset with values of the variables we have information apriori about. This dataset must be of "data.frame" class.

domainChild

An "numeric" array with the range of the child variable.

domainParents

A matrix of class "matrix" with the range of the parent variables or an "numeric" array if there is a only one parent.

One of the inputs and the output of the recursive function "conditional".

conditionalfunction

The output of the function learn.tree.Intervals.

Value

The main function conditionalMethod returns a list with the name of the parents, the different intervals and the fitted functions.

Details

The main function conditionalMethod fits truncated functions for a variable which depends on others variables. The domain of the parent variables is splitted in different intervals and univariate functions are fitted in these ranges. The rest of the described functions are internal ones of the main function.

Examples

Run this code

# NOT RUN {
## Dataset
X <- rnorm(1000)
Y <- rbeta(1000, shape1 = abs(X)/2, shape2 = abs(X)/2)
Z <- rnorm(1000, mean = Y)
data <- data.frame(X = X, Y = Y, Z = Z)

## Conditional Method
parents <- c("X","Y")
child <- "Z"
intervals <- 2

potential <- "MTE"
fMTE <- conditionalMethod(data, nameParents = parents, nameChild = child, 
numIntervals = intervals, POTENTIAL_TYPE = potential)
printConditional(fMTE)

##############################################################################
# }
# NOT RUN {
potential <- "MOP"
fMOP <- conditionalMethod(data, nameParents = parents, nameChild = child,
numIntervals = intervals, POTENTIAL_TYPE = potential, maxParam = 15)
printConditional(fMOP)
# }
# NOT RUN {
##############################################################################

##############################################################################
## Internal functions: Not needed to run #####################################
##############################################################################
# }
# NOT RUN {
domainP <- range(data[,parents])
domainC <- range(data[, child])
t <- conditional(data, nameParents = parents, nameChild = child,
domainParents = domainP, domainChild = domainC, numIntervals = intervals,
mm = NULL, POTENTIAL_TYPE = potential)
printConditional(t)
selection <- select(data, nameParents = parents, nameChild = child,
domainParents = domainP, domainChild = domainC, numIntervals = intervals,
POTENTIAL_TYPE = potential)
parent1 <- selection$parent; parent1
domainParent1 <- range(data[,parent1])
treeParent1 <- learn.tree.Intervals(data, nameParents = parent1,
nameChild = child, domainParents = domainParent1, domainChild = domainC,
numIntervals = intervals, POTENTIAL_TYPE = potential)
BICscoreMoTBF(treeParent1, data, nameParents = parent1, nameChild = child)
# }
# NOT RUN {
###############################################################################
###############################################################################
# }