New titles published weekly — explore the Discussion Publishers catalogue
← Research Center
Mathematics
Open Access
CC-BY

An Introduction to Mathematical Statistics

Fetsje Bijma, Reinie Erné, Marianne A. Jonker, Aad van der Vaart · 2017

Abstract

This book gives an introduction into mathematical statistics.It was written for bachelor students in (business) mathematics, econometrics, or any other subject with a solid mathematical component.We assume that the student already has solid knowledge of probability theory to the extent of a semester course at the same level.In Chapter 1, we give the definition and several examples of a statistical model, the foundation of every statistical procedure.Some techniques from descriptive statistics that can assist in setting up and validating statistical models are discussed in Chapter 2. The following chapters discuss the three main topics in mathematical statistics: estimating, testing, and constructing confidence regions.These subjects are discussed in Chapters 3, 4, and 5, respectively.Next, Chapter 6 provides deeper theoretical insight, in particular into the question under what circumstances and in what sense certain statistical models are mathematically optimal.In Chapter 7, we describe several regression models that are commonly used in practice.The theory from the previous chapters is applied to estimate and test unknown model parameters and give confidence regions for them.Finally, in Chapter 8, we discuss model selection.In that chapter, various criteria are presented that can be used to find the best-fitting model from a collection of (regression) models.Sections and examples marked with a * are more difficult and do not belong to the basic subject matter of mathematical statistics.Every chapter concludes with a summary.In Appendix A, we recall elements from probability theory that are relevant for understanding the subject matter of this book.In Appendix B, we discuss properties of the multivariate normal distribution, which is used in several sections.Appendix C contains tables with values of distribution and quantile functions of several distributions to which we refer in the text.These are meant to be used at home or during problem sessions.In "real life," these tables are no longer used: the computer is faster, more accurate, and easier to use.The statistical package R, for example, contains standard functions for the distribution function, the density function, and the quantile function of all standard distributions.The mathematical style of this book is more informal than that of many mathematics books.Theorems and lemmas are not always proved or may be formulated in an informal manner.The reason is that a pure mathematical treatment is only possible using measure theory, of which we do not assume any knowledge.On the other hand, the relevance and motivation of the theorems are also clear without going into all the details.Each chapter concludes with a case study.It often contains a statistical problem that is answered as well as possible based on the collected data, using the statistical techniques and methods available at that point in the book.The R-code and data of these applications, as well as the data of several case studies described in the book, are available and can be downloaded from the book's webpage at http://www.aup.nl.Though this book includes examples, practice is indispensable to gain insight into the subject matter.The exercises at the end of each chapter include both theoretical and more practically oriented problems.Appendix D contains short answers to most v