rcatbn: creates a pseudo-random numcat bn object

In rebastaba: Handling Bayesian networks

Description

(bn) From a gn object adds on all nodes numcat probability distributions. The parameters for these are pseudo-randomly choosen according to arguments 'nbcg', 'co' and 're'. For each beta distribution, the two parameters will be taken equal. This explains why there is only one interval for each of the two parameters.
Using this function with the same set.seed starting point (and the same request) must produce identical bn.
The gn can be any graph (without cycle) for instance designed by hand or resulting from rgn.

Usage

rcatbn(gn, description=new("des", name=gn@description@name), nbcg=c(3, 5),
	  co=c(1, 1), re=c(3, 3))

Arguments

gn	The starting gn object.
description	The name for the created object (description object).
nbcg	numbers of categories for each node are chosen with equal probabilities within the numbers nbcg[1]:nbcg[2].
co	The first parameter when calling categ3betap is drawn with runif with these two values.
re	The second parameter when calling categ3betap is drawn with runif with these two values.

Value

a bn object comprising only categoric variables

Examples

 rebastaba3k("RESET"); # (only for R checking)
 print(rcatbn(g4n.gn7), quoi="l", qui=9, proba=TRUE);

rebastaba documentation built on May 2, 2019, 5:24 p.m.