Factor Prior — prior_factor • RoBMA

Create priors on factor contrasts.

Usage

prior_factor(
  distribution,
  parameters,
  truncation = list(lower = -Inf, upper = Inf),
  prior_weights = 1,
  contrast = "meandif"
)

Arguments

distribution: character. Prior distribution name.
parameters: list. Distribution parameters.
truncation: list with lower and upper truncation bounds.
prior_weights: numeric prior model weight.
contrast: character. Contrast coding used for factor levels. Common RoBMA model options are "treatment", "meandif", and "orthonormal"; the standalone helper also accepts BayesTools contrast aliases such as "independent".

Value

An object inheriting from prior, prior.factor, and a contrast-specific marker class.

Details

Mean-difference and orthonormal contrasts require vector or multivariate priors. Treatment/dummy and independent contrasts use univariate simple priors per contrast coefficient.