plan0: generates values from one node of a bn

In rebastaba: Handling Bayesian networks

Description

(bn) A vector of N values is extracted from a (most often marginal) node according to different ways to serve as value(s) when further generating a simulation design.
The possibilities depends on the node: root, distribution...
The result depends also on the nature of the variable ( numeric or character).

Usage

plan0(bn, ion, N=5, type="U")

Arguments

bn	The original bn.
ion	indicates the unique node from which values must be returned (iin or name).
N	number of values to extract.
type	The way to extract the values: "U" for equally spaced along the plausible range (numeric) "R" for randomly drawn from the distribution (this applies only for root nodes), "P" for proportional to the distribution (this applies only for numeric root nodes and when the distribution is a standard case where R provides the cumulative distribution. For the moment, only the "U" option is implemented.

Details

Notice that the result can be random, so it is advisable to monitor the pseudo-random seed value. Multivariate nodes are accepted and their variables are considered as independent nodes.

Value

The resulting vector of values (in fact a matrix with as many columns as variables in the node, and N rows).

Examples

 rebastaba3k("RESET"); # (only for R checking)
 plan0(rebastaba.bn2, 1);
 plan0(rebastaba.bn2, 2);

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