bnlearn (version 1.6)

bnboot: Parametric and nonparametric bootstrap of Bayesian networks

Description

Apply a user-specified function to Bayesian networks learned from bootstrap samples of the original data.

Usage

bnboot(data, statistic, R = 200, m = round(2/3 * nrow(data)),
    sim = "ordinary", algorithm, algorithm.args = list(),
    statistic.args = list(), debug = FALSE)

Arguments

data

a data frame, containing the variables in the model.

statistic

a function or a character string (the name of a function) to be applied to each bootstrap replicate.

R

a positive integer, the number of bootstrap replicates.

m

a positive integer, the size of each bootstrap replicate.

sim

a character string indicating the type of simulation required. Possible values are "ordinary" (the default) and "parametric".

algorithm

a character string, the learning algorithm to be applied to the bootstrap replicates. Possible values are gs, iamb, fast.iamb, inter.iamb, mmpc and hc. See

algorithm.args

a list of extra arguments to be passed to the learning algorithm.

statistic.args

a list of extra arguments to be passed to the function specified by statistic.

debug

a boolean value. If TRUE a lot of debugging output is printed; otherwise the function is completely silent.

Value

• A list containing the results of the calls to statistic.

References

Friedman N, Goldszmidt M, Wyner A (1999). "Data Analysis with Bayesian Networks: A Bootstrap Approach". In "UAI '99: Proceedings of the 15th Annual Conference on Uncertainty in Artificial Intelligence", pp. 196-20. Morgan Kaufmann.

See Also

constraint-based algorithms, mmpc, hc, hybrid algorithms.

Examples

data(learning.test)
bnboot(data = learning.test, R = 2, m = 500, algorithm = "gs",
  statistic = arcs)
# [[1]]
#      from to
# <arcs for the first replicate>
#
# [[2]]
#      from to
# <arcs for the second replicate>