Compute residual network

Interpret the error correlations as a residual network model (Epskamp et al. 2017) .

Usage

bmasem_residual_network(object)

Arguments

object

(bmasem) An object of bmasem-class returned by bmasem.

Value

A data.frame containing posterior samples of the partial correlation matrix.

References

Epskamp S, Rhemtulla M, Borsboom D (2017). “Generalized Network Psychometrics: Combining Network and Latent Variable Models.” Psychometrika, 82(4), 904–927. ISSN 1860-0980, doi:10.1007/s11336-017-9557-x , 2024-07-01.

Examples

if (FALSE) { # \dontrun{
model_syntax <- paste("distress =~", paste0("x", 1:14, collapse = " + "))
fit <- bmasem(
  model_syntax,
  sample_cov = Norton13$data, sample_nobs = Norton13$n
)
res_net <- bmasem_residual_network(fit)
p_corr_df <- posterior::summarise_draws(res_net)
n_items <- sqrt(nrow(p_corr_df))
p_corr_mat <- matrix(p_corr_df$mean, n_items)
p_corr_mat
qgraph::qgraph(p_corr_mat)
} # }