CalculatePriorRatio {EDISON}R Documentation

Calculates the network prior ratio.

Description

This function calculates the ratio of the network structure priors for a structure move.

Usage

CalculatePriorRatio(method, q, lambda, network.info)

Arguments

method

Indicates which prior to use: 'poisson' for the standard Poisson prior (no information sharing), 'exp_soft' or 'exp_hard' for the exponential information sharing prior with soft or hard sharing among nodes and 'bino_soft' or 'bino_hard' for the binomial information sharing prior with soft or hard sharing among nodes.

q

Number of nodes in the network.

lambda

Vector of lambda hyperparameter values for each network (needed for the Poisson prior).

network.info

The network information collected using CollectNetworkInfo.

Value

Returns the ratio of the network structure priors for the proposed structure move.

Author(s)

Frank Dondelinger

References

For more information on the network structure priors, see:

Husmeier et al. (2010), "Inter-time segment information sharing for non-homogeneous dynamic Bayesian networks", NIPS.

Dondelinger et al. (2012), "Non-homogeneous dynamic Bayesian networks with Bayesian regularization for inferring gene regulatory networks with gradually time-varying structure", Machine Learning.

See Also

CalculateLikelihoodRatio


[Package EDISON version 1.1.1 Index]