I am a first year student in a PhD program in statistics with near zero statistics background prior to this year. I must say, this is probably the best balance of theory/applications I've come across thus far. The proofs are rigorous but not untouchable to many non-statistics students in the course (ie, epidemiology and related fields requiring a statistics background). Book is thorough and the applications/figures well-motivate the topic area. Definitely a great book to read through and then keep on the shelf.
Content wise, when I first got the book, naturally I paged to the end and legitimately the entire thing looked foreign/illegible to me more or less it seemed so advanced. However, PHoff presents the material in such a way that he provides just enough theory to make the results feel intuitive but not cumbersome, which really helps downstream with your ability to recall and actually APPLY the results within the book. The book basically takes you from zero bayesian background to understanding, coding, and applying fairly rigorous techniques for seemingly cumbersome statistical models that will feel intuitive to you by the end.
Only drawback to the book is that there is no second edition. There are a number of errors throughout the book that definitely detract somewhat from the content. However, if you check PHoff's website, he has noted (as far as I can tell) all of them in annotations by page, which helps this issue a bit.
Also, a "first course in bayesian statistics" as per the preface assumes you have a very solid understanding of fundamentals of probability and statistics (NOT that you know how to use a normal distribution, or other statistical results, you should be familiar with theoretical fundamentals for this book to make sense). Hence it is a first course in bayesian statistics, and not a first course in statistical theory. Particularly, you should definitely be very comfortable with Bayes' rule, and how to manipulate around probability statements using Bayes' rule and definitions of conditional probability fairly seamlessly, before giving the book a try as it assumes you are comfortable with those concepts going in. Also, experience with frequentist inference would be useful for putting Bayesian inference into perspective.
- 语种： 英语
- ISBN: 1441928286
- 条形码: 9781441928283
- 商品尺寸: 15.5 x 1.6 x 23.5 cm
- 商品重量: 440 g
- ASIN: 1441928286
- 用户评分: 分享我的评价