Simulate Datasets from the Generative Model

Generate complete synthetic datasets from a fitted INLAvaan model. For each simulation, a single parameter vector is drawn (from the posterior or prior), and then sample.nobs observations are generated from the model-implied distribution at that parameter value.

Usage

# S4 method for class 'INLAvaan'
simulate(
  object,
  nsim = 1L,
  seed = NULL,
  sample.nobs = NULL,
  prior = FALSE,
  samp_copula = TRUE,
  silent = FALSE,
  ...
)

Arguments

object

An object of class INLAvaan.

nsim

Number of replicate datasets to generate (default 1).

seed

Optional random seed (passed to set.seed()).

sample.nobs

Number of observations per dataset, per group for multigroup models (a single number is used for every group). Defaults to the sample size of the original data.

prior

Logical. When TRUE, parameters are drawn from the prior; when FALSE (default), from the posterior.

samp_copula

Logical. When TRUE (default) and prior = FALSE, posterior parameter draws use the copula method. Ignored when prior = TRUE.

silent

Logical. When TRUE, suppresses the informational message about rejected non-PD draws. Default FALSE.

...

Additional arguments (currently unused).

Value

A list of length nsim. Each element is a data frame with sample.nobs rows and two attributes:

• "truth" — named numeric vector of lavaan-side (x-space, constrained) parameter values used to generate the dataset.

• "truth_theta" — named numeric vector of the corresponding unconstrained (theta-space) parameter values.

Details

This function is designed for tasks that require full replicate datasets from a single parameter draw, such as simulation-based calibration (SBC) and posterior predictive p-values. It differs from sampling() which generates one observation per parameter draw (useful for prior/posterior predictive density overlays).

For each simulation \(s = 1, \ldots, S\):

1. Draw \(\boldsymbol\theta^{(s)}\) from the posterior (or prior).

2. Compute the model-implied covariance \(\boldsymbol\Sigma(\boldsymbol\theta^{(s)})\). If it is not positive-definite, reject and redraw.

3. Generate a dataset of sample.nobs rows from \(N(\boldsymbol\mu(\boldsymbol\theta^{(s)}),\, \boldsymbol\Sigma(\boldsymbol\theta^{(s)}))\).

Parameter draws reuse the same internal machinery as sampling() (sample_params_prior / sample_params_posterior), so the prior specification is consistent.

See also

sampling() for single-observation draws from the predictive distribution (prior/posterior predictive checks).

Examples

utils::data("HolzingerSwineford1939", package = "lavaan")
fit <- acfa("visual =~ x1 + x2 + x3", HolzingerSwineford1939)
#> ℹ Mode finding and Hessian computation.
#> ✔ Posterior mode and Hessian. [57ms]
#> 
#> ℹ Performing VB correction.
#> ✔ VB correction; mean |δ| = 0.285σ. [261ms]
#> 
#> ⠙ Fitting 0/6 skew-normal marginals.
#> ✔ Fit 6/6 skew-normal marginals. [120ms]
#> 
#> ⠙ Posterior sampling and summarising.
#> ⠹ Computing fit indices (PPP/DIC).
#> ✔ Summarise 1000 posterior draws. [496ms]
#> 
#> ℹ Fit measures: PPP, DIC.

# Simulate one replicate dataset from the posterior
sims <- simulate(fit, nsim = 1)
head(sims[[1]])                    # data frame
#>         x1       x2       x3
#> 1 6.097574 6.170981 3.397978
#> 2 5.307746 5.412729 1.416872
#> 3 6.053322 6.835808 1.570789
#> 4 5.115698 5.612655 3.831171
#> 5 5.572679 7.092703 4.118865
#> 6 3.741463 6.247935 1.626214
attr(sims[[1]], "truth")           # true lavaan-side (x-space) parameters
#>     visual=~x2     visual=~x3         x1~~x1         x2~~x2         x3~~x3 
#>      0.8579458      1.3470337      0.9645885      1.1236440      0.5058988 
#> visual~~visual 
#>      0.5378249 
attr(sims[[1]], "truth_theta")     # corresponding unconstrained (theta-space) parameters
#>     visual=~x2     visual=~x3         x1~~x1         x2~~x2         x3~~x3 
#>     0.85794580     1.34703366    -0.03605371     0.11657698    -0.68141870 
#> visual~~visual 
#>    -0.62022225 

# Simulate from the prior (e.g., for SBC)
sims_prior <- simulate(fit, nsim = 5, prior = TRUE)
lapply(sims_prior, nrow)
#> [[1]]
#> [1] 301
#> 
#> [[2]]
#> [1] 301
#> 
#> [[3]]
#> [1] 301
#> 
#> [[4]]
#> [1] 301
#> 
#> [[5]]
#> [1] 301
#>