Assemble a distributional piecewise SEM from fitted drmTMB models

drm_psem() is the piecewiseSEM-style core: you fit each endogenous node yourself with drmTMB::drmTMB() and pass the fitted models. drmSEM extracts the component-labelled graph, validates it as a DAG, and provides path tables, d-separation tests, and simulation-based effects on top.

Usage

drm_psem(
  ...,
  data = NULL,
  covariances = NULL,
  composites = NULL,
  latents = NULL,
  feedback = NULL,
  latent = NULL
)

# S3 method for class 'drm_sem'
summary(object, ...)

Arguments

...: Named fitted drmTMB objects, one per endogenous node. The name is the node identifier used in path queries; predictors are matched to a node by its name or its response variable. A joint bivariate fit may be passed once (unnamed, or named as either response): it is split into the two response nodes and the residual rho12 edge is recorded automatically.
data: The data frame all nodes were fitted to. Defaults to the data of the first node.
covariances: Optional covary() declaration(s) (one object or a list) recording residual (rho12) or higher-level (corpair) covariance edges between responses. These are reported by covariances() and respected by basis_set() / dsep(), but never enter paths() or the effect decomposition.
composites: Optional drm_composite() declaration(s). For drm_psem() the construct column(s) must already be present in the data the nodes were fitted on; the declarations are recorded so loadings() can report them.
latents: Optional drm_latent() declaration(s) or list of them.
feedback: Optional drm_cycle() declaration(s) naming a feedback (cyclic) motif. Cycles are an error unless declared; a declared motif relaxes the topological-order requirement and is reported by cycles(). Node-wise fitting of a declared cycle is inconsistent under simultaneity (a warning is emitted); see docs/design/10-cyclic-feedback.md.
latent: Character vector of marginalised latent vertex names (L_M). When supplied, basis_set() and dsep() generate and test m-separation claims on the implied MAG, conditioning on Richardson & Spirtes (2002) Corollary 5.3 anteriors (S = \\emptyset only). Pairwise claims are jointly licensed by Sadeghi & Lauritzen (2014) Theorem 3 / Lauritzen & Sadeghi (2018) Theorem 4 under a compositional-graphoid assumption (automatic for homoscedastic all-Gaussian nodes; otherwise faithfulness). Conditioned / selection latents are not supported. See docs/design/14-m-separation.md.
object: A drm_sem object.

Value

A drm_sem object.

Details

Fit nodes with control = drmTMB::drm_control(se = TRUE) so that vcov(), Wald intervals, and d-separation refits are available.

References

Shipley B (2009). “Confirmatory Path Analysis in a Generalized Multilevel Context.” Ecology, 90(2), 363–368. doi:10.1890/08-1034.1 .

Lefcheck JS (2016). “piecewiseSEM: Piecewise Structural Equation Modelling in R for Ecology, Evolution, and Systematics.” Methods in Ecology and Evolution, 7(5), 573–579. doi:10.1111/2041-210X.12512 .

Brooks ME, Kristensen K, van Benthem KJ, Magnusson A, Berg CW, Nielsen A, Skaug HJ, Maechler M, Bolker BM (2017). “glmmTMB Balances Speed and Flexibility Among Packages for Zero-Inflated Generalized Linear Mixed Models.” The R Journal, 9(2), 378–400. doi:10.32614/RJ-2017-066 .

Rigby RA, Stasinopoulos DM (2005). “Generalized Additive Models for Location, Scale and Shape.” Journal of the Royal Statistical Society Series C: Applied Statistics, 54(3), 507–554. doi:10.1111/j.1467-9876.2005.00510.x .

Examples

if (FALSE) { # \dontrun{
ctrl <- drmTMB::drm_control(se = TRUE)
size_fit <- drmTMB::drmTMB(
  drmTMB::bf(size ~ temp + habitat, sigma ~ temp),
  family = stats::gaussian(), data = dat, control = ctrl)
abundance_fit <- drmTMB::drmTMB(
  drmTMB::bf(abundance ~ size + temp, zi ~ habitat),
  family = drmTMB::nbinom2(), data = dat, control = ctrl)
sem <- drm_psem(size = size_fit, abundance = abundance_fit, data = dat)
paths(sem)
} # }