estimate8constrainednbn: estimates the parameters of a nbn with equality constraints

In rbmn: Handling Linear Gaussian Bayesian Networks

estimates the parameters of a nbn with equality constraints

Description

Estimations of the parameters of a /nbn/ is done when there are some equality constraints onto the regression coefficients.
Constant terms (mu) and conditional standard deviations (sigma) are supposed independent (that is not constrained with equalities).
Equality constraints are given by sarc, a list of matrices with two columns, indicating each the series of arcs having the same regression coefficient.

Usage

estimate8constrainednbn(nbn, sarc, data, imp=0, nite=10, eps=10^-5)

Arguments

nbn	nbn object.
sarc	List of Matrices with two columns indicating the tails (1rst column) and the heads (2d column) of the arcs having a common parameter. It is checked that these arcs are indeed included in nbn. Nodes must be indicated by their names (not their number).
data	Data frame to be used for the estimation. It must comprise all necessary nodes (not only those involved in sarc but also the remaining parents of sarc[,2]. Usually, all used variables are centred but this is not required.
imp	When 0 nothing displayed. When 1 the number of iterations is displayed. When 2 the successive values of the criterion are also displayed.
nite	Maximum number of iterations.
eps	relative difference in successive scores needed to stop the iterations.

Details

Not linked regression coefficients doesn't require to be included in sarc, the function do it by itself.
The score to use to measure the differences between two successive estimations is not well established (see the code).

Value

The resulting /nbn/ object with the estimated parameters.

Examples

 data(boco);
 print8nbn(rbmn0nbn.05);
 print8nbn(estimate8nbn(rbmn0nbn.05, boco));
 print8nbn(estimate8constrainednbn(rbmn0nbn.05, rbmn0crarc.05, boco));

rbmn documentation built on July 9, 2023, 6:37 p.m.