Simulate data from LTJMM with multivariate normal distribution for random effects

Simulate data from LTJMM with multivariate normal distribution for random effects

# S3 method for class 'ltjmm'
simulate(
  object,
  nsim = object$data$N_obs,
  seed = NULL,
  beta = array(1, c(object$data$N_out, object$data$N_X)),
  gamma = rep(1, object$data$N_out),
  sigma_delta = 1,
  sigma_y = rep(1, object$data$N_out),
  sigma_diag = diag(rep(1, 2 * object$data$N_out - 1)),
  Lcorr = diag(rep(1, 2 * object$data$N_out - 1)),
  delta = NULL,
  ...
)

Arguments

object

a ltjmm object

nsim

number of response vectors to simulate. Defaults to 1.

seed

random seed

beta

fixed effects for covariates

gamma

latent time slope

sigma_delta

standard deviation for latent time

sigma_y

standard deviation for residual variance

sigma_diag

(Cholesky factorization diag_matrix(sigma_diag) * Lcorr ) for random intercepts and slopes

Lcorr

(Cholesky factorization diag_matrix(sigma_diag) * Lcorr) for random intercepts and slopes

delta

vector of latent times. If NULL, latent times delta are simulated by normal(0, sigma_delta).

...

additional optional arguments.

See also

ltjmm