Estimate Marginal Subgroup Treatment Effects — estimate_subgroup

The function uses a counterfactual, marginal approach based on the posterior predictive distribution. It averages over all other covariates to provide robust estimates of subgroup-specific effects.

Usage

estimate_subgroup_effects(
  brms_fit,
  trt_var = NULL,
  data = NULL,
  subgroup_vars = "auto",
  response_type = NULL,
  ndraws = NULL
)

Arguments

brms_fit: A brmsfit object. Fitted model object from fit_brms_model() or run_brms_analysis(). Must contain the necessary attributes for extracting treatment variable and response type information.
trt_var: A character string or NULL. Treatment variable name. If NULL, automatically extracted from model attributes (set by fit_brms_model()). Must be a binary variable coded as 0/1 in the dataset.
data: A data frame or NULL. Dataset used for model fitting. If NULL, automatically extracted from model attributes (set by fit_brms_model()). This dataset is used for generating counterfactual predictions.
subgroup_vars: A character vector or "auto". Subgroup variable names for which to estimate treatment effects. If "auto" (default), automatically detects treatment interaction terms (colon syntax) and random effect grouping factors (pipe syntax) from all formula components (unshrunktermeffect, shprogeffect, shpredeffect).
response_type: A character string or NULL. Outcome type, one of "binary", "count", "continuous", or "survival". If NULL, automatically extracted from model attributes (set by fit_brms_model()). This determines the appropriate scale for effect estimation.
ndraws: An integer or NULL. Number of posterior draws to use for estimation. If NULL (default), all available posterior draws are used. Reducing this can speed up computation at the cost of precision.

Value

list with two named elements:

estimates: tibble where each row represents a subgroup (or "Overall" effect), with columns for Subgroup, Median, CI_Lower, and CI_Upper from posterior distribution
draws: data.frame containing posterior draws for each subgroup

Details

This post-processing function estimates marginal treatment effects for specified subgroups from a fitted brmsfit object.