makebugs: Make BUGS model from fitted DAG

In abn: Modelling Multivariate Data with Additive Bayesian Networks

Make BUGS model from fitted DAG

Description

Make BUGS model from fitted DAG

Usage

makebugs(dag, data.dists, coefs, stderrors)

Arguments

dag	named adjacency matrix representing the DAG. Names correspond to node names.
data.dists	list of node distributions.
coefs	a list named by the node names containing for each element a matrix with the nodes' coefficients.
stderrors	a list named by the node names containing for each element a matrix with the nodes' standard errors

Value

Bugs model returned as stdout.

See Also

simulateAbn gauss_bugs bern_bugs categorical_bugs pois_bugs

Examples

## Prepare data and arguments
mydists <- list(a="gaussian",
                b="multinomial",
                c="binomial",
                d="poisson")
mydag <- matrix(0, 4, 4, byrow = TRUE,
                dimnames = list(c("a", "b", "c", "d"),
                                c("a", "b", "c", "d")))
mydag[2,1] <- mydag[3,2] <- mydag[4,3] <- 1
# plotAbn(mydag, data.dists = mydists)
mycoefs <- list("a"=matrix(-6.883383e-17, byrow = TRUE,
                           dimnames = list(NULL,
                                           "a|intercept")),
                "b"=matrix(c(2.18865, 3.133928, 3.138531, 1.686432, 3.134161, 5.052104),
                           nrow= 1, byrow = TRUE,
                           dimnames = list(c(NULL),
                                      c("b|intercept.2", "b|intercept.3", "b|intercept.4",
                                      "a.2", "a.3", "a.4"))),
                "c"=matrix(c(1.11, 2.22, 3.33, 4.44, 5.55),
                           nrow= 1, byrow = TRUE,
                           dimnames = list(c(NULL),
                                      c("c|intercept", "b1", "b2", "b3", "b4"))),
                "d"=matrix(c(3.33, 4.44),
                           nrow= 1, byrow = TRUE,
                           dimnames = list(c(NULL),
                                      c("d|intercept", "c"))))
mymse <- c("a"=0,"b"=1,"c"=2,"d"=3)
## Make BUGS model
makebugs(dag = mydag, data.dists = mydists, coefs = mycoefs, stderrors = mymse)

abn documentation built on Nov. 3, 2023, 5:08 p.m.