Customer Reviews

Mathematical Statistics and Data Analysis (with CD Data Sets) (Duxbury Advanced)

5 star:		(8)
4 star:		(0)
3 star:		(4)
2 star:		(0)
1 star:		(9)

 

 

This is one of the most readable mathematical statistics textbooks. The level of math used is just right for this course. You need to use is mostly univariate calculus, but partial derivatives may pop up in a few places. You also need to know matrix algebra to read most of the chapter on linear models. All derivations and explanations are clear. The statistics portion of the text starts with sampling theory. In fact, it has more on sampling than I personally wanted to learn, such as sampling from finite populations and stratified sampling. Then it continues with estimation methods and hypothesis testing. There is a very good chapter on descriptive statistics. The later chapters focus on specific models, such as comparing the means of two samples, analysis of variance, categorical analysis, and finally the linear regression. The first chapters summarize probability theory, so the text is very much self-contained even for those with no probability background. There is also a chapter on Bayesian methods, which I haven't read.

The two features that I personally like in this text is that for some models the author presents such modern and sophisticated topics like non-parametric statistical techniques and bootstrap methods for obtaining standard errors. Overall, this is a fine textbook. It gives a solid introduction to statistics for people who need to do statistical analysis on their own and for whom this is their first and last statistics course. This text also gives a nice introduction for students of statistics who plan to move onto more advanced texts. There are one or two places where the derivation is simple but gets a little daunting (I think one of them is one of Wilcoxon's tests). But in the end, you must accept one fact. This is a serious university level applied math text for sophomores/juniors. This book can't be read like a novel, though it is a very readable text as far as mathematical stats goes.

Ignore most of those negative comments from readers many of whom clearly have not mastered their calculus and probability skills before taking a stats course. Undergraduate mathematical statistics is a kind of a course that attracts lots of people from many different majors and math backgrounds. Some take stats only because they have to, not because they want to. In the end, (this may sound arrogant), but the quality and motivation of students in an undergraduate mathematical statistics courses may be very variable. Some probably do not belong in a serious stats course. Hence, the flood of negative reviews for this text.

I work in a large corporation in a technical job. My primary motivation for buying this text book was to apply statistics to practical work-related problems. I loved this book.

I was also amazed to see all the negative reviews. My theory for the negative reviews is that many students have no conception what a real world application of statistics is all about. They think that if they get a job in industry; the statistics problems they encounter involves nothing beyond picking up their favorite statistics book, finding an exact solved example of the problem in the text, copy the formual from the example and then plug in some numbers and you are done!

This book is an introductory-level statistics textbook for people who are learning statistics for the first time, and who know some math but are not math wizards.

I am outright shocked at the low reviews given to this book. I tend to be highly critical of statistics textbooks, especially ones at this level. In my opinion, this book does an outstanding job. It balances making the material accessible with going into considerable depth, and it integrates mathematical theory with an emphasis on data analysis. The title of this book says it all: "Mathematical Statistics and Data Analysis". There are few other books that emphasize both theory and practice. This book certainly lives up to its title in this respect.

Perhaps some of the bad reviews are from frustrated students who are in classes that are moving too quickly for their level. Just because this is an excellent book doesn't mean that every professor will use it in an appropriate way. The exercises in this book can be tough and I think it would be possible to easily overwhelm students by assigning too many problems or problems that are too difficult. This book is only really useful if you are able to explore it at a slower pace, reading it and reflecting on it.

---

This book is not concise but it is not overly wordy. Rice is an excellent writer. The expanded discussion communicates aspects of the subject that are often overlooked, and helps build the students' intuition. This book is always readable and it is clearly written to be read and understood. In addition to communicating basic concepts, the book also explores numerous practical and philosophical considerations. For a book at such an introductory level, this text is remarkably deep.

One of the most attractive aspects of this book is that it is relatively easy to skip around in it. More advanced students or students with some exposure to probability will necessarily want to skip many of the early chapters. Unlike some of the more advanced texts (such as the Casella and Berger) where the chapters depend on and reference each other in a rather rigid fashion, this book is more flexible and thus can appeal to students with diverse backgrounds.

My last bit of praise is the integration of data analysis with mathematical theory. Distributions are introduced with a rich discussion of where, how, and why they arise in practice. Data analysis techniques, tests, and abstract definitions alike are introduced in such a way that they are rooted both to the abstract theory and to consideration of applications: this book is one of the rare texts that fully bridges the gap between the real world and the abstract models used to describe it.

I would recommend this book for a wide variety of different uses. This would make an outstanding textbook and it could easily be used for more than one course, or for courses at more than one level. It is a useful book to have on the shelf as a basic reference, and it is also very useful for self-study. My only complaint is that more books in the field of statistics are not written in the spirit and style of this one. In particular, I would like to see authors of more advanced books examine this book's strengths and use them to enhance their own writing. Or perhaps John Rice ought to try his hand at writing a higher-level text; I know I would be eager to see what he could produce!

I own and have read countless statistics books, from the general to the specialized. Rice's text is simply the most comprehensive, and he does it in far fewer pages than many competitors. Rice includes many theoretical aspects which you'll only find in places like Parzen's MODERN PROBABILITY THEORY AND ITS APPLICATIONS. But best of all, Rice has countless numbers of archetypical examples and problems, ones which serve as frames for real problems that need solving, and provide a rich world for thinking about What Probability And Statistical Ideas Really Mean.

I cannot recommend this book highly enough.

Indeed, I'm surprised its Amazon rating is so poor. That's simply a reflection of the reviewers, to my mind.

My learning in statistics follows a funny order. Before receiving fundamental training in math stat, I took an introductory stochastic processes and graduate econometrics. I use Rice's book for a math stat course this summer and find this book a valuable reference!

This book helps me synthesize what I've learnt from stochastic processes and economoetrics. On top of that, it provides very useful techniques that will assit empirical modelers to perform estimation of probability models and dereviation of random variables. This book is not advanced enough for stat PhD students but definitely a must-read for non-stat majors who want to get a deep understanding of statistical sciences.

I personally obtain ideas from this book to solve problems confronted in model-development and data-manipulation.

The only thing I would complain about this book is its price.

This is an absolutely fantastic book, written by one of the best professors I've ever had for a technical class. I took Prof. Rice's course, Stat 135, at UC Berkeley and it was a really great and illuminating experience. The book, the lectures, and the course overall, cemented my decision to become a Statistics major. The book is technical, but very accessible, and contains applications ranging from Astrophysics to Jane Austen to Finance to Biology and more. It is ideal for self study and exceedingly interesting to just sit down and read. The difficulty level of the problems is fairly distributed such that slow learners have a chance to acclimate themselves to the material, and the quicker ones of us can look to the later problems and be sufficiently challenged.

Don't believe the reviews here or the ones on [...]. They're just outright wrong. This is a great book written by a great professor. If it's over your head, then maybe you don't realize it is written for an upper division Mathematical Statistics course.

I thought this was too theoretical for an introduction to statistics, but it was for a class.

I was asked to buy this book for an introductory course in Statistics. The authors makes a lot of assumptions about prior knowledge and does not explain topics thoroughly.

A number of subjects in this book involve procedures which could be given as recipes, followed by explanations. Instead, the author talks a lot, trying to lead you through some kind of reasoning for what will follow, and never follows up with the recipe.

I understand that students tend to copy the recipe and forget about the rest. But personally, I would rather the students AT LEAST learn the recipe, and be able to understand where it comes from after. Using this text, they will get neither.

Another peeve: most of the figures in this book serve no purpose. The point they are often meant to illustrate is glossed over using numbers; however, unless there is a live demonstration of the computer calculation of these numbers and the consequent figures, NO ONE will pay attention to them.

One example of a section I had to supplement heavily was the beginning of Chapter 9, where the most important details of using the Neyman-Pearson Lemma are left unstated. Rice leaves it to the student to glean the details from a single example.

I find it interesting that the postive reviews for this book seem to be from other authors of stats books providing an apology for one of their colleagues! But if you look at the majority of the student reviews, you see that this book is not highly rated. As a student myself using this book for a grad level stats course, I also have to give it a very low score. I found this book to be extremely difficult to understand. I've had other undergrad stats courses but even so I found this text difficult to follow. Things are not explained well and I did not think there were enough examples worked out in detail. I found myself having to resort to using other stats books to actually read and learn concepts and only used this text to reference the problems that my instructor assigned for homework. Kind of a waste for a book that costs $100+!