Compute Data-Adaptive Priors for Stan Models — compute_dynamic

Inspects preprocessed standard curve data and produces all hyperprior locations and scales needed by the Stan model data {} block.

Usage

compute_dynamic_priors(
  data,
  response_variable,
  fixed_a = NULL,
  model_family = "logistic4"
)

Arguments

data: Data frame of preprocessed standards with concentration and response columns (already on the fitting scale).
response_variable: Character. Response column name.
fixed_a: Numeric or NULL. If non-NULL, the prior on a is tightened to a soft constraint centered on this value (on the fitting scale).
model_family: Character. One of the curveRcore model names. Default "logistic4".

Value

Named list of hyperprior values matching the Stan data block, including prior_log_sigma0_mu, prior_log_sigma0_sigma, prior_log_sigma_slope_mu, and prior_log_sigma_slope_sigma for the heteroscedastic noise path.