Customer Reviews

Numerical Analysis: Mathematics of Scientific Computing

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

 

 

I teach a course in numerical methods every other year and use Maple as the platform for the computer solution of problems. At this time, I am using "Numerical Methods: Third edition" by Faires and Burden and am quite happy with it. However, I am always trolling for better textbooks so that it is premise with which I examined this book.

The level of rigor is on the higher end of the scale, there are many more formal theorems and proofs than is found in most of the other numerical methods textbooks. Rather than using code in Maple or Mathematica, the algorithms are expressed in a lower level pseudocode. It has the appearance of a programming language; however some of the operations are expressed in mathematical form for brevity. This use of the mathematical syntax generally makes them easier to understand than if they were expanded out using a programming language. There are many exercises at the end of the sections, including a group meant for computer solution.

I will not be adopting this book for my class, the level of rigor is a bit higher than my students can easily digest and quite frankly, I am hooked on using Maple as the computer platform. Nevertheless, it serves as a valuable secondary reference, where when necessary, I can look through a detailed proof of a topic that I am about to present.

The book was a major disappointment. I am glad that I did not purchase it for my class, but instead borrowed it. The ordering of topics and emphasis choices never seemed to make sense to me. The layout throughout most of the text is like one long, run-on sentence. The underlying structure of numerical analysis never developed and I was left swimming in meaningless details while the basics were short-changed by an over abundance of specialized algorithms. Perhaps the text's curriculum could be saved by a capable professor, but alas my professor was just as scatter-brained as the text. More pictures would also have been helpful. A replacement text I recommend, which covers the first, matrix theory portion of this book, is David S. Watkins' Fundamentals of Matrix Computations.

I think this book is lucently written and explains various aspects of numerical analysis in great detail. The proofs are stated in an understandable way and algorithms are presented clearly and in such a way that it is easy to implement them in the programming language of one's choice.

This book is very good. It is bit dense for studying (lot of material/problems etc). You wont repent it if you buy it.

If you are a graduate student then this book might appeal to you. But if you are actually trying to use the book for some purposeful end, it will probably disappoint.

The first problem is that it uses lots and lots of mathematical symbols, as well as plenty of abstruse matrix transformations. If you are studying for your PhD then you probably will have no trouble.

The next problem is that the book is pretty much page after page of solid proofs. The proofs are essentially meaningless; they are presented without context and are rarely used for anything except building other proofs. In the places where numerical examples are presented, they are so brief and cryptic as to be unintelligible.

Believe it or not, the same authors wrote another book, Numerical Mathematics and Computing - it covers almost exactly the same material, but in a more down-to-earth way. If you are determined to read Numerical Analysis, then I would highly recommend that you read the other book first. Instead of having copious proofs, the other book spends its time reviewing the mathematical methods required to apply the material, as well as reviewing the notation.

Honestly, this has turned out to be a horrible book. Particularly in the disconnect that exists between the text and the problem sets. Very few of the examples are useful to read through because they are trivially simple, while the problem sets seem to take particular delight in finding the hardest tricks to be solve-able.

I would highly suggest that anyone that purchases this book, already know what they are trying to learn, or have an excellent teacher that can fill in the gaps.

Bought this as a textbook for my class. The book is pretty well-organized and easy to read. I like it. I was reading some other books of Numerical Analysis, and they are pretty old, maybe that's why they are less organized than this one.

I bought this book for my college and it was excellent