ABSTRACT
This introduction presents an overview of the key concepts discussed in the subsequent chapters of this book. The book aims to bring to a wide audience an encompassing picture of the promise of this new class of Chain Event Graph (CEG) models which can be used for analysing many common classes of problems in a coherent, comprehensive and transparent manner. It outlines central results from the theory of Bayesian networks (BNs), repeating very briefly how this popular class of models has been used in the past. The book opens the door for the use of CEGs in probabilistic inference. It explains how to embed a given problem description into a CEG and how to, vice versa, read modelling assumptions directly from a CEG. The book moves on to exploring the properties of discrete and parametric statistical models represented by a CEG. It presents how to use CEGs in statistical inference.
