Changes in version 1.5.1 (2025-11-19) NEW FEATURES - Added preference_index() to compute a model-based preference index from Plackett–Luce worth parameters. The function reports, for each item: - expected top probability (%), - expected bottom probability (%), - net winning probability (%), and - normalized worth. - Added rank_tricot2() — a redesigned ranking parser for tricot data. The function: - handles ties and missing values more robustly, - enforces stricter validation of A–B–C syntax, - provides clearer error messages, and - returns rankings in a format consistent with Plackett–Luce models. IMPROVEMENTS - Updated internal utilities in rank_numeric() to interface with rank_tricot2() for improved performance and error handling. - Minor internal cleanup (no user-visible behavioural changes). BUG FIXES - None reported for this release. Changes in version 1.4 (2024-12-05) IMPROVEMENTS - manage factor levels in plot.pltree() - introduces the functions kendallTau_permute() and kendallTau_bootstrap() - adds S3 method to plot PlackettLuce objects using plot.PlackettLuce() - adds cassava to the datasets - adds cassava consumer testing data and new vignette - improves readability of common bean vignette - coefficients from worth_map() are extracted using stats::coefficients() instead of psychotools::itempar() - update the bean vignette Changes in version 1.3 (2024-05-27) IMPROVEMENTS - Implements the function resample() to re-sample Plackett-Luce coefficients - Adds standard errors in regret() - Adds pairwise_probs() to compare worth parameters following the Luce's Axiom BUG FIXES - Fix the computation of maximized regret in regret(). Using sum(x^2) instead of sum(x)^2 Changes in version 1.2 IMPROVEMENTS - Implement anova() method for PlackettLuce models - Implement likelihood-ratio test for PlackettLuce models. Function likelihood_ratio() Changes in version 1.0 (2023-04-21) IMPROVEMENTS - kendallTau() returns Z-value and p-values Changes in behaviour - Functions rank_binomial() and rank_paircomp() are renamed as set_binomialfreq() and set_paircomp(), respectively Changes in version 0.7 (2023-02-26) IMPROVEMENTS - Adds argument validate.rankings to rank_tricot() to help in identifying possible issues in input tricot ranking like NA, ties or letters different than A, B or C - Adds argument average to kendallTau() to allow returning the kendall correlation for each entry in a matrix and PlackettLuce ranking - Adds argument na.omit to kendallTau() to remove entries with NA - Code is written using = instead of <- to avoid typing stress BUG FIXES - Fixes an issue in plot.pltree() when multi-comparison is required but multicompView is not loaded. Changes in version 0.5 (2022-09-23) IMPROVEMENTS - Fix issues in CRAN cmd checks Changes in version 0.4 (2022-06-28) IMPROVEMENTS - Add new functions reliability(), btpermute(), pseudo_rank(), kendallW() - Split plot.pltree() to build branches and panels independently Changes in version 0.3 (2022-03-08) First version on CRAN Changes in version 0.2.5.9000 IMPROVEMENTS - add argument qve to skip quasi-variance to plot coefficients in plot_tree() - rank_numeric() deals with ids of class characters - argument ref can be used in multicompPL() to select the reference item - adjust scale in the plot() method for summarise_agreement() using the argument scales, where scales = 100 is to set the scale from 0-100 and scales = 1 for a scale from 0-1 - plot method for multcompPL() BUG FIXES - Fix an issue in rank_binomial() in assigning the ids when argument disaggregate = TRUE - Fix an issue in setting up the permuted formulas in btpermute(), with the fix n.formula = n.vars + 1 Changes in version 0.2.2 NEW FEATURES - forward selection is added with forward() - btpermute() is added for model selection with BTm() - abbreviation method in plot methods of summarise_* functions IMPROVEMENTS - No user visible improvements Changes in version 0.1.9 CHANGES IN BEHAVIOUR - Functions rainfall(), temperature(), ETo() and GDD() migrates to climatrends - Change license to MIT - Change logo Changes in version 0.1.8 IMPROVEMENTS - Write documentation in good practice. Avoid calling the entire package but specific functions. CHANGES IN BEHAVIOUR - argument "ascending" in rank_numeric() in restricted to floating point numbers. Integer values are ranked as provided by input. Changes in version 0.1.7 NEW FEATURES - Classify main functions into families - rainfall() and temperature() now allows for time series indices CHANGES IN BEHAVIOUR - Rename functions agreement(), dominance(), favourite(), and victories() with the tag summarise and add it into summarise functions. No warking message provided. New functions are summarise_agreement(), summarise_dominance(), summarise_favourite(), and summarise_victories() - Argument 'index' is removed from rainfall() and temperature(), all indices are given by default Changes in version 0.1.6 NEW FEATURES - .get_timeseries() now implements the Euclidean method to search for its nearest neighbour in the nasapower output. This optmises the function and reduce dependencies - ranking functions now implements internal processess to decode ranks and reduce dependencies - pseudoR2() as a S3 method for objects of class "glm", "gnm", "bttree" and "pltree" - kendallTau() as S3 method for objects of class "numeric", "matrix", "data.frame", "rankings", "grouped_rankings" Changes in version 0.1.4 NEW FEATURES - Function to summarise results from pairwise comparisons, agreement(), dominance(), favourite() and victories() - Increased test coverage Changes in version 0.1.3 - Successfull migration from ClimMobTools to gosset Changes in version 0.1.2 NEW FEATURES - Migrating functions from ClimMobTools to gosset CHANGES IN BEHAVIOUR - predict() method for crossvalidation and rank_decimal() deprecated - rank_numeric() and rank_tricot() replaces rank_PL(). NO warning message added Changes in version 0.1.1 NEW FEATURES - predict() method for crossvalidation objects - pseudoR2() for object of class "pltree" extracts null loglik from PlackettLuce - replace to_rankings() by rank_PL(). Warning message added - replace num2rank() by rank_decimal(). No warning message added - replace to_paircomp() by rank_paircomp(). No warning message added Changes in version 0.1.0 - GitHub-only release of prototype package.