BinoHyperMove {EDISON}

Makes a binomial hyperparameter move.

Description

This function proposes a move for one of the hyperparameters of the binomial prior, calculates the acceptance probability and accepts the move accordingly.

Usage

BinoHyperMove(network.info, node.sharing, GLOBvar)

Arguments

network.info

The collected network information obtained using CollectNetworkInfo.

node.sharing

Which type of node sharing is used, either 'soft' or 'hard' sharing.

GLOBvar

Global variables of the MCMC.

Value

Returns a list with elements:

move

The move type (in this case, 2).

move.made

1 if the move was proposed, 0 otherwise.

network.info

The network information, including the new hyperparameters if the move was accepted.

accept

Whether the move was accepted or not.

Author(s)

Frank Dondelinger

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