Customer Reviews

Aspects of Statistical Inference (Wiley Series in Probability and Statistics)

5 star:		(0)
4 star:		(0)
3 star:		(1)
2 star:		(0)
1 star:		(0)

 

 

This book is very data-oriented and very application-oriented. It requires a great deal of background, and although the prose is clear and easy to read, the mathematics can be extraordinarily difficult to get through. The emphasis is always on academic rigour: this book explores the practical and philosophical questions about how to put statistical theory to use in areas of application. The book does not emphasize theory, but rather, assumes that the student already understands the theory inside and out.

I very strongly emphasize that the background requirements necessary to understand this book are great. I would recommend that students be very comfortable with probability theory, mathematical statistics, linear algebra, and have at least one course in abstract math and at least one course in some sort of applied math, before attempting to use this book. Although this book certainly does not focus on mathematics, the mathematics used in this book are raw and ugly. When equations appear, they can be very complex.

The mathematical aspects of statistical inference are simply not developed in this book. Theorems are stated and not proven. There is a brief appendix, with a long list of mathematical "facts", all without proof or explanation.

I'm not quite sure what the authors were thinking when they wrote this book, but it does not stand on its own. I think the authors would have done well to either provide more discussion of the mathematics, or to simplify the mathematics used--either choice would make the book more self-contained. In some respects, this book is complemented by the widely used Casella and Berger (which I think has the exact opposite flaws as this book--it focuses exclusively on the theory, at the expense of practical considerations and applications).