Model-Based Clustering for Multivariate Partial Ranking Data
Simulate a sample of ISR(pi,mu)
Getter method for rankclust output
Summary function.
Convert data
Change the representation of a rank
Criteria estimation
Cayley distance between two ranks
Hamming distance between two ranks
Kendall distance between two ranks
Spearman distance between two ranks
Convert data storage
Khi2 test
Kullback-Leibler divergence
Constructor of Output class
Probability computation
Constructor of Rankclust class
Model-based clustering for multivariate partial ranking
Model-Based Clustering for Multivariate Partial Ranking Data
Show function.
Implementation of a model-based clustering algorithm for ranking data (C. Biernacki, J. Jacques (2013) <doi:10.1016/j.csda.2012.08.008>). Multivariate rankings as well as partial rankings are taken into account. This algorithm is based on an extension of the Insertion Sorting Rank (ISR) model for ranking data, which is a meaningful and effective model parametrized by a position parameter (the modal ranking, quoted by mu) and a dispersion parameter (quoted by pi). The heterogeneity of the rank population is modelled by a mixture of ISR, whereas conditional independence assumption is considered for multivariate rankings.