Component effects from a cpaic fit

Usage

component_effects(object, newdata = NULL, ...)

# S3 method for class 'cpaic_mlnmr'
component_effects(
  object,
  newdata = NULL,
  level = 0.95,
  estimand = "average_conditional_link",
  ...
)

Arguments

object: A fitted cpaic object (cpaic_bridge, cpaic_maic, cpaic_stc, or cpaic_mlnmr).
newdata: For cmlnmr() fits: a one-row data frame giving the target population's effect-modifier values, at which the component effects beta + Gamma x are reported. With newdata = NULL (default) the component main effects beta are returned; these are the effects at the covariate origin and are not population-adjusted.
...: Passed to methods.
level: Credible level for the component-effect intervals (default 0.95), for cmlnmr() fits.
estimand: The only implemented cML-NMR component-effect estimand is "average_conditional_link".

Value

A data frame of component effects (estimate, se, CI, p-value). Components that the design cannot identify are returned as NA.