Statistical Packages

  • labeltodendro with Anthony Davison and Arpit Chaudhary

    • Convert labels or tables to a dendrogram. This package offers a dendrogram representation of series of labels as especially needed in Markov Chain Monte Carlo clustering. If you have a dendrogram in your mind you can easily put series of meaningful labels in a matrix and heights in a vector, then convert them to a dendrogram object.

  • bclust with Anthony Davison

    • The package builds a dendrogram with log posterior as a natural distance defined by the model. It is also capable to computing Bayesian discrimination probabilities equivalent to the implemented Bayesian clustering. Spike-and-Slab models are adopted in a way to be able to produce an importance measure for clustering and discriminant variables. The method works properly for data with small sample size and high dimensions and offers a fully automatic method.

  • baybi (under development)

    • This package clusters both variables and subjects and at the same time says which variables are important for clustering subjects and which subjects are important for clustering variables.

    [HOME]