Customer Reviews

Wavelet Methods for Time Series Analysis (Cambridge Series in Statistical and Probabilistic Mathematics)

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

 

 

The purpose of this text is to introduce wavelet methodology targeted toward applications in time series analysis. Emphasis is placed on the discrete wavelet transform and other discrete wavelets. There are many analogies to the spectral analysis of time series that the authors make. The authors do not assume knowledge of Fourier analysis and cover the relevent theory in Chapter 2. Just as the frequency domain methods decompose the variance of a time series into frequency components so to do wavelets decompose the variance according to scales. The authors take great pains in demonstrating this with several wavelets including the DWT and MODWT. Fast transform methods exist for wavelets as well as for Fourier series and they are apparently even simpler and faster for wavelets.
The text tries to be self-contained and the authors start out at a very elementary level. They emphasize applications and practical aspects of wavelets.

Few wavelet books offer such a deep insight into wavelet methods
for time series analysts. It goes very deep within the applied
side of wavelet analysis on real time series while not compromising
the mathematical side. A lot of books and papers insist too much
on the mathematical side of wavelets while this one provides just
the right balance between rigor and practical insight. If your
interest does not lie in maths but just in wavelets as a tool,
stick to this one.

As a bonus, it contains a lot of exercices along with answers
at the end of the book...a very good textbook indeed...

As far as I am concerned,anything Don Percival writes about the topic of wavelets - or digital signal processing - is well worth reading. I was struggling to understand wavelets before I ran across a paper by Percival (and cohorts from the University of Washington) on interpreting remote sensing sea ice data using wavelets. After reading the very clear and compact description of wavelet analysis the light bulb clicked on for me. This book is a must read for anyone interested in wavelets from beginners to experienced practicioners.

Well written book. It would be nice to have some "MatLab" problems. Excellent motivational or explanatory examples.

I gave it three stars because most reviewers are far to liberal with their stars. To me 3 stars means worth buying, 4 means a must buy, and 5 means it'll be a classic for a long time.

I equate this book to Mallet's "A Wavelet Tour of Signal Processing". The subject matter is slightly different, but both are very readable.

This is an excellent self-contained book on wavelet theory and its applications. I am a physicist and I have found in this book not only a clear explanation of the fundamentals but a suitable exposition of the advanced topics. This is a very self-study text!!

The book has interesting features but presented to me several drawbacks. Though it tries to be self-contained in introducing the wavelet subject, the resultin theoretical structure is poor. and general properties of wavelets remain scattered for all over the book, leaving some gaps. Perhaps my opinion is biased by being used to classical introductions stressed on Multi Resolution Analysis. Moreover, the habit of carrying on calculation with explicit indexes, quoting intermediate results and unusual notations (e.g. denoting a function as x(.) instead of x(t) in order not to bind to a specificic variable) clutter the book with cumbersome expressions, even if in that form they can be straightforwardly rendered in computer routines. I just wonder how people without a previous knowledge about wavelet methods can have dealt with the textbook.

This textbook has the advantage of being exceedingly thorough and detailed. This, unfortunately, is also its main drawback. Professional mathematicians will probably enjoy the extensive derivations. Applied engineers, statisticians and programmers, however, may find this book to be rather long-winded, and too complex for context. As a mathematical treatise, this work is unsurpassed. However, it is definitely not a user-friendly programming manual.

This book offers a concise highly mathematical accurate analysis treatment of modern wavelet theory. It covers all essential material and should be suitable for those involved in Digital Signal Processing.