Customer Reviews

Statistics: A Bayesian Perspective

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

 

 

This is an excellent introductory text designed for a first course in statistics. It covers all the topics that are typically in a first course. However, all other texts at this level take the frequentist approach to inference. A few may have sections that introduce Bayesian ideas but the Bayesian approach is a paradigm for statistical inference and as such the approach should be incorporated in all statistical topics. Berry shows that this can be done without the student having to know calculus. To understand Bayesian methods the student mainly has to know that posterior probability = likelihood x prior probability. Berry provides a good list of references for those who want to pursue more advanced topics.
This book is unique. It demonstrate that statistics can be taught from the Bayesian approach in the very beginnning. This is much like what Noether did when he wrote an introductory text in statistics taking a strict nonparametric approach.

The text is loaded with exercises and the exposition is very clear. There are many useful and entertaining diagrams. Many examples are taken from real medical problems. Medicine is an area in which Berry has done a great deal of consulting and his experience shows in his examples. This should be the text to turn to if you want an introduction to the subject. If you know the basics and want more advanced treatment go to the references mentioned in Berry's preface.

This book completely fulfills its goals, one of which is not to be a definitive reference book. It provides a friendly, entertaining introduction into statistics from a Bayesian perspective.

This is a truly clear and thoughtful introduction to Bayesian statistics.Nothing is taken for granted as the author leads you through examples and concepts. This was my first introduction to Bayesian statistics, and Berry makes it seem so much more reasonable and closer to real research/real life than the artifice involved in other approaches.

As a primer on Bayesian statistics for an intermediate or advanced statistics course,
Donald Berry has written a fascinating an insightful guide for the reader who enjoys
algebraic calculation and mathematical proofs. However its claims that it is
designed as "an introductory textbook on statistics" is where the reader may
be deceived. In the preface, Berry writes "mathematical level of this book is
minimal with only an exposure to high school algebra expected".

There are many excellent textbooks introducing the subject of statistics,
and aimed at people with only high school algebraic experience - this book is not
one of them however. Instructors should read this text carefully, and completely,
before even considering assigning it as reading in an introductory course.

The book itself is incredibly dense, and often the problems contained within
require extensive knowledge of statistical calculations and probability theory.
Even as a graduate student who adores statistical techniques, this is heavy going.
That is not to say that it is completely inaccessible to the reader, and would
not be an appropriate introductory text for graduate study. The reader may find
it frustrating, however, going back and learning statistics from first principles,
but will gain some insight into Bayesian probability theory.

Additionally, the book requires the use of Minitab, and does not evaluate other alternate
computer methods of calculation. If you're happy to do calculations by hand, this book
may still be of use, but at a minimum it needs to be revised and computer calculations
discussed. There are many free or online Bayesian calculators available, which you
should explore before falling back to the books hand calculations.

Excellent for self study. I was able to follow everything up to chapter 11 completely unaided.

It is not too useful for people beyond college level. Not as a reference book.