stages_bj: Backward joining of stages

In stagedtrees: Staged Event Trees

Backward joining of stages

Description

Join stages from more complex to simpler models using a distance and a threshold value.

Usage

stages_bj(
  object = NULL,
  distance = "kullback",
  thr = 0.1,
  scope = NULL,
  ignore = object$name_unobserved,
  trace = 0
)

Arguments

object	an object of class sevt with fitted probabilities and data, as returned by full or sevt_fit.
distance	character, see details.
thr	the threshold for joining stages
scope	names of variables that should be considered for the optimization.
ignore	vector of stages which will be ignored and left untouched, by default the name of the unobserved stages stored in object$name_unobserved.
trace	if >0 increasingly amount of info is printed (via message).

Details

For each variable in the model stages are joined iteratively. At each iteration the two stages with minimum distance are selected and joined if their distance is less than thr.

Available distances are: manhattan (manhattan), euclidean (euclidean), Renyi divergence (reny), Kullback-Liebler (kullback), total-variation (totvar), squared Hellinger (hellinger), Bhattacharyya (bhatt), Chan-Darwiche (chandarw). See also probdist.

Value

The final staged event tree obtained.

Examples

DD <- generate_xor_dataset(n = 5, N = 1000)
model <- stages_bj(full(DD, lambda = 1), trace = 2)
summary(model)

stagedtrees documentation built on April 29, 2022, 1:06 a.m.