YGGDRASIL - A Statistical Package for Learning Split Models
This work addresses the need for more sophisticated modeling of conditional independencies in specific contexts, primarily for statisticians and data analysts working with categorical data, but it appears incremental as an extension of existing graphical models.
The paper introduces a statistical framework for split models, which extend graphical models to handle context-specific independence structures in contingency tables, and presents the YGGDRASIL software package for learning these models from data.
There are two main objectives of this paper. The first is to present a statistical framework for models with context specific independence structures, i.e., conditional independences holding only for sepcific values of the conditioning variables. This framework is constituted by the class of split models. Split models are extension of graphical models for contigency tables and allow for a more sophisticiated modelling than graphical models. The treatment of split models include estimation, representation and a Markov property for reading off those independencies holding in a specific context. The second objective is to present a software package named YGGDRASIL which is designed for statistical inference in split models, i.e., for learning such models on the basis of data.