Paul Damien, Petros Dellaportas, Nicholas G Polson and David A Stephens (Eds)
Oxford University Press, £95.00
Bayesian statistical methods (the estimation of an unknown phenomenon of interest) are increasing in popularity and finding new practical applications in many fields of biological health sciences, as well as engineering, environmental sciences, business and economics, and social sciences, among others.
This book (which is published in honour of distinguished British statistician Sir Adrian F M Smith) claims to be a clear, concise description of all the major ideas in Bayesian statistics. I suspect that the terms 'clear' and 'concise' might be somewhat relative if you are not familiar with some very advanced mathematic concepts. This book is not simple and it assumes a fair amount of knowledge of what Bayes's theorem is and how it works.
Assuming you have the appropriate background in statistics, this book is a useful introduction to the field. It has an interesting format. There are 12 parts and in each case (bar the first chapter), journal-style papers follow an introduction, each providing further details and example applications of a specific technique or application. The book is therefore both a textbook and compilation of papers that provides details on most of the past and present advances in the field.
In summary, this is an excellent book for its intended audience – statisticians who wish to learn Bayesian methods – but the rest of us may be better starting off with a more basic text.
Dr Oliver Jones CBiol MSB