Dynamic Mixed Models for Familial Longitudinal Data
The book focuses on the analysis of familial and longitudinal data by means of dynamic mixed models. These data are correlated as (a) the responses of the members of a particular family share a common random family effect and (b) the repeated responses of the same individual are not independent. These familial and longitudinal correlation structures play a crucial role in the analysis of this type of data and greatly condition the model estimates and the rest of the statistical inferences. The book consists of eleven chapters. The first exhaustively lists the most relevant antecedents on familiar and longitudinal models in the last three decades. Chapters 2 and 3 give an overview of the analysis of longitudinal data by linear models (fixed and mixed, respectively). The rest of the chapters are divided in four pairs and each pair focusses on the study of count data and binary data, respectively, using distinct kind of models. In particular, the following: familial models (chapters 4 and 5), longitudinal models (chapters 6 and 7), longitudinal mixed models (in the following pair of chapters), and finally familial longitudinal models (in the last pair). The different models are discussed in depth with a special care for the technicalities behind the countless amounts of different correlation models. This means that the statistical background of the potential readers of the book needs to be rather advanced and the reading becomes occasionally arduous. This drawback is alleviated by the fact that the book also broaches the analysis of a considerable number of real life data sets, mainly within the context of biostatistics and econometrics. Researchers of these two areas along with applied statistics undoubtedly constitute the main target of the book.