Skip to contents

DBN HMM Model

Source: R/models_hmm.R
dbn_hmm.Rd

Regime-switching DBN for large-scale networks (100+ actors, 50+ time steps)

Usage

dbn_hmm(
  Y,
  family = c("ordinal", "gaussian", "binary"),
  R = 3,
  nscan = 10000,
  burn = 1000,
  odens = 1,
  delta = rep(1, R),
  seed = 6886,
  verbose = TRUE,
  time_thin = 1,
  previous = NULL,
  init = NULL,
  symmetric = FALSE,
  ...
)

Arguments

Y

Data array (nodes x nodes x relations x time)

family

Data family (ordinal, gaussian, or binary)

R

Number of regimes

nscan

Number of iterations of the Markov chain (beyond burn-in)

burn

Burn-in for the Markov chain

odens

Output density for the Markov chain

delta

Dirichlet prior for transition matrix

seed

Random seed

verbose

Print progress

time_thin

Save every nth time point, when 1, save all time points

previous

Previous dbn_hmm results to continue from (optional)

init

List of initial values: S, A_list, B_list, Pi, sigma2_proc, tau_A2, tau_B2, g2, pi0 (optional)

symmetric

Logical. If TRUE, enforce B = A for each regime. Default: FALSE.

...

Additional arguments

Value

List containing MCMC results

See also

dbn for the main dispatcher, param_summary for posterior summaries

Examples

# \donttest{
sim <- simulate_hmm_dbn(n = 8, time = 10, R = 2, seed = 1)
fit <- dbn_hmm(sim$Y, R = 2, nscan = 200, burn = 100, verbose = FALSE)
# }