computeRho4: Calculate proposal frequencies for changepoint moves.
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
lambda
Hyperparameter controlling the number of hidden states.
Value
Vector containing the proposal frequencies for the different
changepoint moves.
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.