The PL–LBM jointly partitions ranking rows and items. Each pair of latent groups has its own positive strength.
Usage
pl_lbm_model(
latent_strength = .default_latent_strength(),
ranking_clustering = gnedin_prior(gnedin_hyperparameter = 0.5),
item_clustering = gnedin_prior(gnedin_hyperparameter = 0.5),
identifiability = pl_identifiability(),
ability_prior = NULL,
ranking_prior = NULL,
item_prior = NULL
)Arguments
- latent_strength
Gamma prior for positive cell strengths.
- ranking_clustering
Prior on latent ranking groups.
- item_clustering
Prior on latent item blocks.
- identifiability
PL reporting constraint.
- ability_prior
Deprecated alias for
latent_strength.- ranking_prior
Deprecated alias for
ranking_clustering.- item_prior
Deprecated alias for
item_clustering.