Per-parameter interpretations for a symbolized model

Returns the per-coefficient reading tibble carried by a symbolized_model as a reader-friendly table. Each row corresponds to one fixed-effect coefficient and pairs it with the prose readings already templated by symbolize(): a link-scale reading, a natural-scale reading, an optional variance-scale reading (for scale submodels), and a biological reading.

The renderer is a pure consumer of x$interpretation. All prose was substituted from inst/extdata/interpretation-templates.csv at extract time; this surface only displays it.

Usage

parameter_interpretation(
  x,
  scale = c("all", "link", "natural", "variance", "biological"),
  ...
)

Arguments

x

A symbolized_model.

scale

One of "all" (default), "link", "natural", "variance", or "biological". "all" returns every reading column; the other values return just the identifier columns (submodel, term_label, coefficient_role, estimate) plus the matching single reading column.

...

Reserved for future use.

Value

A tibble (S3 class symbolizer_interpretation) with submodel, term_label, coefficient_role, estimate, and one or more reading columns. The "scale" attribute records which scale was requested.

Examples

# parameter_interpretation(symbolize(glm(am ~ wt, binomial(), data = mtcars)))