biased-dot-bn-dot-fit: Random network with a biased degree distribution In ddgraph: Distinguish direct and indirect interactions with Graphical Modelling

Description

A version of random.bn.fit which generates a graph based on degree distribution and beta distribution for probabilities

Usage

 1 biased.bn.fit(nodes, beta.est, in.degree.distr, bn.graph)

Arguments

 nodes character vector of node names beta.est the beta distribution parameters for different degrees of a node. Should be a list where [] corresponds to 2-dimenstional contingency table (i.e. one parent, one output). It contains a data.frame with columns shape1, shape2 for the beta distribution, and rows are degrees of freedom (in this case 2, when P(Out=0|Parent=0) and P(Out=0|Parent=1)) in.degree.distr a vector with degree distribution for all the nodes in the network (names are ignored, and degree is randomly sampled from this vector) bn.graph if the graph structure is already available, then the graph structure in object of class "bn"

Value

a list of two elements: bn - a bn object which contains the structure and bn.fit - a bn.fit object with filled in conditional probabilities

Examples

 1 2 3 4 5 6 # nodes, conditional probability distribution, an indegree distribution nodes = letters[1:5] beta.est = list(data.frame(shape1=2,shape2=3), data.frame(shape1=c(2,4), shape2=c(5,2)), data.frame(shape1=c(1,2,3,4), shape2=c(3,2,1,2))) in.degree.distr = c(0, 1, 1, 2, 2) # make a random graph using these parameters biased.bn.fit(nodes, beta.est, in.degree.distr)

ddgraph documentation built on Nov. 17, 2017, 10:50 a.m.