A Practical Guide to Heavy Tails: Statistical Techniques and Applications

Rating:
Summary: Nothing more than a compilation of papers !
Review: Even if some of the papers whithin the book are certainly important contributions to their field, my impression is that it does not justify paying that price for extended versions of the papers. The book by itself is merely a compilation of the papers but NOTHING MORE ! If you are interested, just get the papers on the web and read them, but don't buy this book !

I hoped i would find some unifying ideas in the book, or even remarks in the extended versions of the papers but nothing... I really wonder how it got published. The only useful information is the extended references for every paper (hence i gave 1 star) ...but it's not worth the price of the book !

I have been extremely disappointed because i'm used to very good work from M. Taqqu, including numerous outstanding papers, but this book is the worst i've ever bought. It does not deserves the name "book".

Rating:
Summary: AN INTERESTING BOOK ON HEAVY-TAILED DISTRIBUTIONS
Review: In principle, I must confess that I am astonished by the low rating given by the other reviewer! The book is a very good one, emphasizing topics that have not appeared in any other book so far (as, for example, applications of heavy-tailed distributions to financial modeling).

The reason for which I did not give 5 stars is because I am a computer-network specialist, hence I was expecting to see more networking applications than financial ones.

Nevertheless the mathematical content of the book is at a high level; I can assure everyone about it because my first degree is a first-class B.Sc. in Mathematics.

My current job has to do with the design of IP-based multimedia networks and, in fact, I am one of the very few persons (three at most) in the Balkans and probably the only one in Greece who applies the concepts of self-similarity, long-range dependence and heavy-tailed distributions to the design of real communication networks for everyday use. Hence I contend that my opinion matters.

I found particularly helpful the last chapter, which describes numerical techniques. In fact, this is the first book on self-similarity and long-range dependence to address such issues.

I strongly recommend the book for applied mathematicians and financial-modeling experts, while it is quite an interesting reference for network-performance experts. I also recommend it strongly for research students in any of these disciplines.