Skip to contents

Simulate from HMM DBN Model

Source: R/simulate.R
simulate_hmm_dbn.Rd

Generates data from a regime-switching HMM DBN

Usage

simulate_hmm_dbn(
  n = 30,
  n_col = n,
  p = 2,
  time = 50,
  R = 3,
  sigma2 = 0.5,
  tau_A2 = 0.2,
  tau_B2 = 0.2,
  transition_prob = 0.8,
  stickiness = NULL,
  seed = NULL,
  return_truth = TRUE,
  symmetric = FALSE
)

Arguments

n

Number of sender actors

n_col

Number of receiver actors (default: n)

p

Number of relation types

time

Number of time points

R

Number of regimes

sigma2

Innovation variance

tau_A2

Prior variance for regime A matrices

tau_B2

Prior variance for regime B matrices

transition_prob

Diagonal transition probability

stickiness

Deprecated alias for transition_prob

seed

Random seed

return_truth

Return true latent states and parameters

symmetric

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

Value

List containing simulated data and true parameters

See also

dbn for model fitting, simulate_test_data for quick test data

Examples

sim <- simulate_hmm_dbn(n = 8, p = 1, time = 10, R = 2, seed = 6886)
dim(sim$Y)
#> [1]  8  8  1 10
table(sim$S)
#> 
#> 1 2 
#> 4 6