ABSTRACT
It is clear that these two classes refer to statistical modelling and inference. In the same paper, for the first time, Fisher coined the term ‘likelihood’ explicitly and contrasted it with ‘probability’, two “radically distinct concepts [that] have been confused under the name of ‘probability’....” Likelihood is a key concept in both modelling and inference, and throughout this book we shall rely greatly on this concept. This book aims to extend classical likelihood inferences to general classes of models, including unobservable random variables, because it is important to have a deep understanding of existing likelihood inferences. This chapter summarizes all the classical likelihood concepts from Pawitan (2001) that we need in this book for extensions; occasionally, for more details, we refer the reader to that book.
