bpcs v1.2.2

• Fixing a bug in the solve_ties option

Release 1.2.1

bpcs 1.2.1

• Moving to cmdstanr instead of rstan.
  • This will allow us to fix some bugs and tweaks that were not optimal.
  • Now we can have faster installations and let cmdstan compile the models.
  • There will be some additional time to compile the models for the first time but that is only the first time we use it
  • We can now remove the errors from ubsan-clang in CRAN which apparently is a lot of trial and error to solve and not supporting tools from CRAN for identifying that
• The interface of the bpcs will remain (practically) the same
• new function to retrieve the posterior distribution of the parameters get_parameters_posterior
• alias to retrieve the summary data frame of the parameters get_parameters_df
• Removed dependency on coda
• Now we can specify the probability mass in the parameters and in summary
• rstan and shinystan are now optional
• Fix problems with multiple clusters in the posterior predictive function
• Ability to add credibility mass and choose HDPI or credible intervals in print and in summary
• New function check_convergence_diagnostics for HMC diagnostics. This is printed as default in print but not in summary

Release 1.1

bpcs 1.1.0

• Possibility to add up to 3 intercept random effects (hopefully you will never need more than that)
• Model for subject predictors (see the example on the paper)
• Make predictions of submodels with the model_type option (see the example on the paper)
• Some small bug fixes

In the next release, we will remove support from rstan and move towards cmdstanr

Minor improvements and publication functions

• removed ties_pred from models that do not have ties and from the stan models.
• fixed predict for models with ties, so we return a vector y_pred with 0, 1, 2 and not separate as now
• removed posterior distributions from the get_rank_of_players and get_probabilities. Now we have new functions to obtain the data frame or the posterior distributions separately. The posterior is now returned as matrix
• Probabilities table is now optional in the summary function
• New functions to get the probabilities for specific data get_probabilities_newdata_df and get_probabilities_newdata_posterior
• Publication ready functions for
  • plots: get_parameters_plot function and a thin S3 plot wrapper for the same function. Plots are default to APA.
  • tables: Functions for publication tables: get_parameter_table, get_probabilities_table and get_rank_of_players_table
• get_hpdi_parameters became get_parameters and the user specify if credible intervals or hpdi
• Added ties to expand_aggregated_data.
• We can get now both credible and HPD intervals in get_parameters. n_eff and Rhat are also now possible to add and remove from this df
• Added functions to save and load bpc models

First official release

Features of the bpcs package

• Bayesian computation of different variations of the Bradley-Terry
  (including with home advantage, random effects and the generalized
  model).
• Bayesian computation of different variations of the Davidson model
  to handle ties in the contest (including with home advantage, random
  effects and the generalized model).
• Accepts a column with the results of the contest or the scores for
  each player.
• Customize a normal prior distribution for every parameter.
• Compute HDP interval for every parameter with the
  get_hpdi_parameters function
• Compute rank of the players with the get_rank_of_players function.
• Compute all the probability combinations for one player beating the
  other with the get_probabilities function.
• Convert aggregated tables of results into long format (one contest
  per row) with the expand_aggregated_data.
• Obtain the posterior distribution for every parameter of the model
  with the get_sample_posterior function.
• Easy predictions using the predict function.
• We do not reinforce any table or plotting library! Results are
  returned as data frames for easier plotting and creating tables
• We reinforce the need to manually specify the model to be used.

Models available

• Bradley-Terry (bt) (Bradley and Terry 1952)
• Davidson model (davidson) for handling ties (Davidson 1970)

Options to add to the models:

• Order effect (-ordereffect). E.g. for home advantage (Davidson and
  Beaver 1977)
• Generalized models (-generalized). When we have contestant
  specific predictors (Springall 1973)
• Intercept random effects (-U). For example, to compensate
  clustering or repeated measures (Böckenholt 2001)

E.g.:

• Simple BT model: bt
• Davidson model with random effects: davidson-U
• Generalized BT model with order effect: bt-generalized-ordereffect