categ3betap: compute a p-variable categ distribution

In rebastaba: Handling Bayesian networks

Description

(ba) from beta distributions generates the joint probability table for p (=length(n)) categoric variables coming from the discretization of the betas (approximately equal spaced discretized).

Usage

categ3betap(co=c(1, 1), re=c(2, 2), n=c(3, 4, 5), xnames=rbsb.cha0)

Arguments

co	) First parameter when calling categ3beta2.
re	) Second parameter when calling categ3beta2.
n	) Number of classes to use for the p variables. Its length provides the number of variables.
xnames	dimnames of the resulting probability array. The default generates some standard names.

Details

The central idea is the product of pairs of beta obtained from categ3beta2. For instance, in case of p=3, if P1(i,j) is associated to the first two variables, P2(i,k) is associated to the first and last variables \& P3(j,k) is associated to the last two variables, then the joint proba is P(i,j,k) propto P1(i,j)P2(i,k)P3(j,k).

Value

a probability array of dim n

Examples

 rebastaba3k("RESET"); # needed only for R checking, to be forgotten
 categ3betap();

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