sampleParms: Sample initial parameters for the MCMC simulation.
Description
This function samples the initial hyperparameters and parameters that are
needed for the MCMC simulation.
Usage
sampleParms(X, GLOBvar, HYPERvar, s_init = NULL, options)
Arguments
GLOBvar
Global variables of the MCMC simulation.
HYPERvar
Hyperparameter variables.
s_init
Initial number of changepoints.
Value
Returns a list with elements: - E
- The initial changepoint
vector.
- S
- The intial networks structure.
- B
- The initial
regression parameters.
- Sig2
- The inital sigma squared variances.
- betas
- The intial hyperparameters for the exponential information
sharing prior.
- hyper_params
- The initial hyperparameters for the
binomial information sharing prior.
References
For more information about the parameters and hyperparameters,
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.