Author: Ngai Hang Chan
Publisher: Wiley
ISBN: 978-0470414354

Click Here To Purchase Analysis of Financial Time Series (Wiley Series in Probability and Statistics)

In the past few years there have been several changes in the financial landscape as well as developments in using time series techniques for financial modeling. Time Series pplications to Finance with R and S-Plus aims to highlight several of these standard as well as non-standard techniques applied in finance using S-Plus and R as statistical analysis tools.
The initial part of the book deals with standard time series techniques, starting with simple descriptive techniques for trends and seasonality of data. Key definitions and results in probability theory and stochastic processes provide a strong theoretical foundation for progressing into advanced topics in later chapters.

Commonly used models such as the Auto Regressive Moving Average (ARMA) are introduced but with a very effective way. One example of the effectiveness of the book is a clear Q&A section on the duality between causality and stationarity.


The treatment of the subject covers both time and frequency domain techniques. Estimation in the time domain focuses on ARMA models and standard issues such as order selection. In explaining forecasting, the author starts with simple forecasting and moves on to the Box-Jenkins approach and connects the techniques and implementation with an implementation on T-Bills.


On
the frequency domain analysis, there is a good treatment of spectral representation theorem and periodogram. Examples in S-Plus and R are provided through the text and there is a chapter devoted just for implementation examples as the book gets into more advanced topics.


Handling non-stationarity in mean and variance is described followed by techniques for heteroscedasticity including ARCH and GARCH. The book covers practical aspects of these models including estimation and testing of the models and shows the implementation of a practical example for forex rates.


Building further on multivariate time series, Vector Auto Regression models are described with an example of inference with VaR. State Space modeling including representation, Kalman recursion, stochastic volatility models are explained clearly followed by a Kalman filtering example of the term structure.

Multivariate GARCH modeling is explained using an example of modeling volatility in higher dimensions. The chapter also explore various features of S+ in fitting a multivariate GARCH.

In discussing co-integration and common trends, the focus is on the basic concept and foundations of the structure of the co-integrated system.  There is a solid example using spot and futures.


The discussion on Markov Chain Monte Carlo finishes with Bayesian inference, basic concepts including the Metropolis Hastings algorithm and Gibbs sampling, and a case study that analyzes the impact of jumps on the Dow Jones Index.


One of the key attractions of the book is that each chapter is very focused and covers an important aspect of modern time series analysis succinctly. The chapters have the right blend of theory and practical implementation and are easy to understand.

Time Series: Applications to Finance with R and S-Plus is a great complement to Analysis of Financial Time Series (Wiley) by Ruey S Tsay, reviewed previously in this site, progressing from fundamental techniques to more advanced techniques in a clear and succinct way. The book not only does a great job of highlighting the subtleties of using time series techniques in finance, but also gives an in-depth understanding in a structure that is geared towards the practitioner.

Click Here To Purchase Analysis of Financial Time Series (Wiley Series in Probability and Statistics)