EDISON (version 1.1.1)

BinoHyperRatio: Calculates the MH ratio of the binomial prior.

Description

This function calculates the ratio of the binomial information sharing prior with the proposed new hyperparameter values, and the binomial prior with the current hyperparameter values.

Usage

BinoHyperRatio(params.proposed, changed, node.sharing, network.info)

Arguments

params.proposed

The new proposed hyperparameter values for the binomial prior.

changed

Gives the index of the parameter that has changed.

node.sharing

Type of information sharing among nodes: 'soft' or 'hard'.

network.info

The network information as collected by CollectNetworkInfo.

Value

This function returns a number greater than zero which represents the ratio of binomial priors.

References

For information about the binomial information sharing prior, 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

BinoHyperMove