computeRho4: Calculate proposal frequencies for changepoint moves.

In EDISON: Network Reconstruction and Changepoint Detection

Description

This function calculates the frequency at which each of the different changepoint moves is proposed. For the poisson network structure prior, this ensures that the proposal frequency is equal to the prior probability.

Usage

computeRho4(k, kmin, kmax, c, lambda)

Arguments

k	The number of hidden states.
kmin	Minimum number of hidden states.
kmax	Maximum number of hidden states
c	Parameter.
lambda	Hyperparameter controlling the number of hidden states.

Value

Vector containing the proposal frequencies for the different changepoint moves.

Author(s)

Sophie Lebre

References

For more information about the hyperparameters and the functional form of the likelihood, see:

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

EDISON documentation built on May 2, 2019, 2:39 a.m.