| Index | Topics |
| rbn {bnlearn} | R Documentation |
Simulate random samples from a given Bayesian network
Description
Simulate random samples from a given Bayesian network.
Usage
rbn(x, n = 1, ..., debug = FALSE)
Arguments
x |
an object of class |
n |
a positive integer giving the number of observations to generate. |
... |
additional arguments for the parameter estimation procedure, see again |
debug |
a boolean value. If |
Details
rbn() implements forward/logic sampling: values for the root nodes are sampled from their
(unconditional) distribution, then those of their children conditional on the respective parent sets. This
is done iteratively until values have been sampled for all nodes.
If x contains NA parameter estimates (because of unobserved discrete parent
configurations in the data the parameters were learned from), rbn() will produce samples that
contain NAs when those parent configurations appear in the simulated samples. See
bn.fit for details on how to make sure bn.fit objects
contain no NA parameter estimates.
Value
A data frame with the same structure as the data originally used to fit the parameters of the Bayesian network.
Author(s)
Marco Scutari
References
Korb K, Nicholson AE (2010). Bayesian Artificial Intelligence. Chapman & Hall/CRC, 2nd edition.
See Also
Examples
data(learning.test)
dag = hc(learning.test)
fitted = bn.fit(dag, learning.test)
rbn(fitted, 5)
| Index | Topics |
