Introduction to empirical processes and semiparametric inference (Q2465368): Difference between revisions

Empirical processes and inference in semiparametric statistical models are two mutually stimulating research areas in mathematical statistics. This book is an introduction to what is commonly called the modern theory of empirical processes -- empirical processes indexed by classes of functions -- and to semiparametric inference, and the interplay between both fields. Part I consists of four chapters: Introduction; An overview of empirical processes; Overview of semiparametric inference; and Case studies. I. This part covers basic notions and results without providing a detailed technical treatment in order to acquaint the reader with a somewhat global picture of the two fields. Here and in the two subsequent parts of the book the chapters on case studies discuss applications of the general theory to specific models, as, for example, the Cox model or partly linear logistic models. Part II is devoted to the development of the theory of empirical processes with the chapters: Introduction to empirical processes; Preliminaries for empirical processes; Stochastic convergence; Empirical process methods; Entropy calculations; Bootstrapping empirical processes; Additional empirical process results; The functional delta method; Z-estimators; M-estimators; and Case studies. II. As becomes clear from the chapter headings, in this part all important aspects of empirical process theory are presented in a linear and systematic fashion. Finally, Part III discusses semiparametric inference in the chapters: Introduction to semiparametric inference; Preliminaries for semiparametric inference; Semiparametric models and efficiency; Efficient inference for finite-dimensional parameters; Efficient inference for infinite-dimensional parameters; Semiparametric M-estimation; and Case studies. III. The book closes with a list of more than 140 references, an extensive and hence useful list of symbols, and both an author and a subject index. This is clearly intended to be a book for the novice in empirical process theory and semiparametric inference. As is stated in the author's preface, and becomes very clear from the notes at the end of each chapter, the book owes quite a bit to the books of \textit{A. van der Vaart} and \textit{J.A. Wellner} [Weak convergence and empirical processes. With applications to statistics. NY: Springer (1996; Zbl 0862.60002)], and \textit{P.J. Bickel}, \textit{C.A.J. Klassen}, \textit{Y. Ritov} and \textit{J.A. Wellner} [Efficient and adaptive estimation for semiparametric models. Baltimore, MD: Johns Hopkins Univ. Press (1993; Zbl 0786.62001), NY: Springer (1998; Zbl 0894.62005)]. In fact, it may be considered as a preparatory text to these more advanced monographs. In the chapters on ``Preliminaries'', for example, the reader will find prerequisites from the theories of metric spaces, Banach spaces and Hilbert spaces. The main material is presented in a clearly arranged and logical order. However, the book is not fully self-contained. A number of times the reader is referred to other sources, mainly the two books mentioned before, for the proofs of essential facts like, for example, Prokhorov's theorem in the chapter on stochastic convergence. On the other hand, some new results are also presented; for example, the continuous mapping results in the chapter on bootstrapping empirical processes. The book will be useful to anybody who wants to learn about the modern theory of empirical processes and semiparametric inference.