Package index • BTSBM: Bayesian paired-comparison and ranking models

All functions

ATP_2000_2025: ATP Paired-Comparison Panels, 2000–2025
as_bt_data(): Create validated Bradley–Terry pairwise data
as_rankings() as_pl_data(): Create validated Plackett–Luce ranking data
block_winprob_table_from_lambdas(): Block-level Bradley-Terry win probabilities from player strengths
bt_model(): Define a simple Bradley–Terry model
bt_sbm_model(): Define a Bradley–Terry stochastic block model
clean_players_names(): Format player/item names as "Surname F."
clustering_prior(): Choose a prior for a latent partition
compare_bt_models_loo(): Compare BT models with Pareto-smoothed importance sampling LOO
compute_expected_wins_rank_posterior(): Posterior rank summaries from player-strength draws
dirichlet_process_prior(): Dirichlet-process prior for a latent partition
eurovision_toy(): Eurovision-inspired toy ranking data
exploratory_adjacency(): Adjacency heatmap (wins per matches)
finite_partition(): Define a finite-Dirichlet prior for a latent partition
fit_btsbm(): Fit a Bayesian Bradley–Terry or Plackett–Luce model
gibbs_bt_sbm(): Gibbs sampler for the Bradley–Terry Stochastic Block Model (BT–SBM)
gibbs_bt_simple(): Simple Bradley–Terry Gibbs sampler (no clustering)
gnedin_K_mean(): Expected number of clusters under the Gnedin prior
gnedin_K_var(): Variance of the number of clusters under the Gnedin prior
gnedin_prior(): Gnedin prior for a latent partition
implied_ability(): Compute implied item abilities from a fit
lambda_to_theta(): Map \(\lambda\) to Bradley–Terry \(\theta = \lambda_i / (\lambda_i + \lambda_j)\)
latent_strength() gamma_ability(): Define a Gamma prior for positive latent strengths
log_lik(): Compute pointwise log likelihoods from a BTSBM fit
loo_btsbm(): Run Pareto-smoothed importance-sampling LOO for a fit
make_bt_cluster_loo(): Log-likelihood matrix for the BT–SBM (clustered) model
make_bt_simple_loo(): Log-likelihood matrix for the simple Bradley–Terry model
mcmc_control(): Set MCMC controls for a BTSBM fit
mcmc_diagnostics(): Summarise MCMC mixing diagnostics
partition_moves(): Set partition-move controls
partition_prior(): Legacy partition-prior interface
pitman_yor_prior(): Pitman–Yor prior for a latent partition
pl_identifiability(): Set the identifiable representation for Plackett–Luce abilities
pl_lbm_model(): Define a Plackett–Luce latent block model
pl_lbm_toy(): Structured Plackett–Luce latent-block-model toy data
pl_mixture_model(): Define a Plackett–Luce ranking-mixture model
pl_model(): Define a simple Plackett–Luce model
pl_sbm_model(): Define a Plackett–Luce stochastic block model
plot_assignment_probabilities(): Figure 4 plotting function Assignment probability heatmap (BT-SBM)
plot_block_adjacency(): Figure 3 plotting function Block-ordered adjacency heatmap (BT-SBM)
plot_lambda_uncertainty(): Figure 5 plotting function Lambda uncertainty plot (per player)
plot_mcmc_traces(): Plot saved MCMC traces
plot_pairwise_outcomes(): Plot observed pairwise win counts
plot_posterior_similarity(): Plot the posterior similarity of items or rankings
plot_rank_intervals(): Plot posterior rank intervals
plot_ranking_positions(): Plot how often items appear at each ranking position
plot_strength_summary(): Plot posterior item strengths and their uncertainty
posterior_similarity(): Compute posterior similarity for an inferred partition
posterior_strength(): Summarise posterior item strengths
pretty_table_K_distribution(): Format a block-count posterior distribution as a pretty- table
print(<btsbm_fit>): Print a BTSBM fit
relabel_by_lambda(): Relabel partitions by decreasing \(\lambda\), compute point estimates, and quantify uncertainty via credible balls
sample_from_BTSBM(): Sample a Bradley–Terry Stochastic Block Model (BT-SBM) tournament
strength_summary(): Summarise posterior item strengths with credible intervals
summary(<btsbm_fit>): Summarise a BTSBM fit
sushi_toy(): Sushi-inspired toy ranking data
tennis_toy(): Tennis-league toy paired-comparison data