Trade in Multivariate Time Series Analysis: with R and Financial Applications (Wiley Series in Probability and Statistics) for an Amazon Gift Card of up to £18.75, which you can then spend on millions of items across the site. Trade-in values may vary (terms apply). Learn more
An accessible guide to the multivariate time series tools used in numerous real-world applications Multivariate Time Series Analysis: With R and Financial Applications is the much anticipated sequel coming from one of the most influential and prominent experts on the topic of time series. Through a fundamental balance of theory and methodology, the book supplies readers with a comprehensible approach to financial econometric models and their applications to real-world empirical research. Differing from the traditional approach to multivariate time series, the book focuses on reader comprehension by emphasizing structural specification, which results in simplified parsimonious VARMA modeling. Multivariate Time Series Analysis: With R and Financial Applications utilizes the freely available R software package to explore complex data and illustrate related computation and analyses. Featuring the techniques and methodology of multivariate linear time series, stationary VAR models, VARMA time series and models, unit-root process, factor models, and factor-augmented VAR models, the book includes: Over 300 examples and exercises to reinforce the presented content User-friendly R subroutines and research presented throughout to demonstrate modern applications Numerous datasets and subroutines to provide readers with a deeper understanding of the material Multivariate Time Series Analysis is an ideal textbook for graduate-level courses on time series and quantitative finance and upper-undergraduate level statistics courses in time series. The book is also an indispensable reference for researchers and practitioners in business, finance, and econometrics.
About the Author
RUEY S. TSAY, PhD, is H.G.B. Alexander Professor of Econometrics and Statistics at The University of Chicago Booth School of Business. He has written over 125 published articles in the areas of business and economic forecasting, data analysis, risk management, and process control. A Fellow of the American Statistical Association, the Institute of Mathematical Statistics, and Academia Sinica, Dr. Tsay is author of Analysis of Financial Time Series, Third Edition and An Introduction to Analysis of Financial Data with R, and coauthor of A Course in Time Series Analysis, all published by Wiley.
This text is a good reference for a broad range of topics in the analysis of multivariate time series. It covers not only proven mainstream methods (VARMA, cointegration, PCA) but also delves into some cutting-edge techniques such as those used in volatility modelling. The coverage across topics it not uniform, and the author freely admits the depth of coverage has been influenced by his own preferences and understanding of the various topics. Therefore it should come as no surprise that the text is heavily focused on VARMA modelling.
The order by which topics are presented is reasonable. I like that the author starts right off with the more general vector/matrix approach and does not waste time introducing the "special case" posed by univariate time series, since this book is going to be very difficult for a student who has not already been exposed to the univariate case. Indeed, a level of familiarity with advanced mathematical topics (such as infinite-order matrix polynomials) is presumed. I also liked that proofs occur mostly at the end of chapters, and that the author maintains a web page with a listing of errata.
While this is an excellent reference for a variety of techniques, like many of these kinds of books, it does not offer much on the subject of model selection -- that is, the process of choosing between competing techniques, nor much advice on how to determine whether a particular technique might be appropriate or inappropriate in a given setting.
Some reviewers have complained about being turned off by the heavy use of mathematical notation, some of which may seem foreign to some readers. However, I think this is unavoidable when covering advanced topics such as these. To his credit, the author mostly uses a consistent set of notation throughout the book, and is mostly willing to dispense with mathematical purity whenever it would hinder readability. For example, I was thankful that he uses the same symbol for a random variable and its realization, as using separate symbols would have been unnecessarily complicated in this setting.
2 of 2 people found the following review helpful
Detailed Analysis of Multivariate Time Series (MTS)13 April 2014
Multivariate Time Series Analysis by Ruey Tsay provides a thorough mathematical analysis of time series and how the open source statistical package R can be used for analysis. The author provides an Add-On package, MTS (Multivariate Time Series) which be installed in R and used for analysis. One great feature of the book is the authors's web page which provides all the R code used in the book and allows the reader a fair amount of practice.
Overall, the book is mathematically dense and challenging to follow. I certainly had a tough time following the numerous equations and terminology. In a future edition, the author may want to include more conceptual descriptions and less equations. The book would also more useful if the author would augment the relatively scarce R program code included in the text, even though he does an excellent job in the website associated with the book.
I would not recommended this book for a casual user who just wants to get an overview of the topic and be able to use the R code to solve specific problems of interest. For maximum benefit, the reader will need significant investment in time to allow carrying out the exercises in R.
Overall, this book is best suited for a user with some familiarity in the mathematical foundations of multivariate time series analysis and wants to better understand advanced topics, and how to use R for their analysis.
1 of 1 people found the following review helpful
Another essential reference from Prof. Ruey Tsay20 Feb. 2014
Professor Ruey Tsay is the author of multiple editions of book on time series analysis, most recently An Introduction to Analysis of Financial Data with R. Most people who work with financial data or who have taken a graduate course in time series analysis have a copy of one of Prof. Tsay's books. The previous books focus largely on univariate time series. Prof. Tsay's new book, Multivariate Time Series Analysis focuses on multiple time series that have some kind of cross sectional relationship. Time series with multivariate or co-integrated distributes occur in finance, which accounts for the sub-title.
One of the things that I appreciated about the book is that the proofs are kept to appendix sections at the end of the chapters. I found the level of math approachable. Most of the mathematics is discrete math (summations), without any advanced calculus. Because the book deals with multivariate time series, linear algebra is used through out the book, but again, the level is reasonable. An advanced undergraduate or anyone who has taken applied math graduate courses should have not problem with the math.
Except for some of the notation... Few people read books like this from front to back. Generally there's something that you would like to understand and you read the associated chapters. Jumping into a chapter, I found some of the notation difficult to understand. For example, Prof. Tsay uses B to indicate the back-shift operator (e.g., B(x[t]) = x[t-1]) (see page 19). I have not seen this notation so jumping into a chapter I kept wondering what this B was that I kept seeing in many equations. Some math texts have a page or two that summarizes the local notation. That would have been nice here.
The book covers a variety of topics and in many cases I found the explanation briefer than I would have wished. My favorite time series book is the now dated Modeling Financial Time Series with S-PLUS, Springer (2006) (Prof. Zivot was one of my professors in graduate school). A search will turn this book up in PDF form, so you don't have to buy it. Prof. Zivot's approach works better for me. He introduces a topic and then develops examples, with more discussion. Multivariate Time Series Analysis does provide R examples, but it was difficult for me to understand the jump from theory to example in many cases without pouring over the text several times. One example of this is the coverage of Principle Component Analysis (PCA) for factor models. PCA is introduced and applied to factor models in a few chapters. For me PCA analysis is something that would have been better covered with its own chapter.
Once I got used to Prof. Tsay's notation I found the book easier to understand. Multivariate Time Series Analysis is an important reference and it will join my earlier copy of Prof. Tsay's Analysis of Financial Time Series on my reference shelf.
While many of the concepts in this book are much more advanced than the low-level probability and statistics courses I took in college, I was surprised to find that I was able to comprehend several sections with little difficulty. The author does an excellent job of presenting the information with helpful graphs, charts, and clear and concise writing. This is obviously a book designed to be used in a classroom, so some of the concise sections would undoubtedly be easier to understand with a professor's guidance. It would be an easy book to Study, however, because the facts are presented so clearly.
Students, if this book is assigned to your class - get ready for a tough semester, but don't be frightened by the book. There's little to no fluff and thankfully the writing is well laid out so that some of it actually makes sense.
Professors, if you're thinking about assigning this book, please do your students a favor and offer plenty of supplemental notes. Tsay's book is concise, but the brevity makes many sections very difficult to 'teach yourself'. It's easy to find yourself lost in the rapid-fire information.
Of course, that should be a given, based on the title and publisher. While Wiley has a few more elementary stats texts, most of their books target advanced statisticians, and address their subjects from a very math-heavy, theoretical perspective. This book is no exception to that rule. I'd suggest at least a masters degree in statistics, or equivalent experience.
The R code is helpful, and accompanies most major sections of the book to facilitate analyses.
The author has at least a couple more basic books on financial and time series analysis that may be a better choice for those less comfortable with the fundamentals:
Analysis of Financial Time Series
An Introduction to Analysis of Financial Data with R