Reviewer Manoj
Rengarajan holds a Master of Financial Engineering - University of
California, Berkeley and he works in the investment management
industry and specializes in providing economic and investment outlook
and strategy for global equity and government bond markets. He has an
educational background in financial engineering, business, and
engineering, and professional interests include business,
finance, economics, technology and related areas.
Time Series Analysis & Forecasting by
Example is an introduction to time series analysis for students who
have some background in statistics. The book starts with a focus on
serial correlation which is one of the hurdles for students who
transition to time series analysis from basic
statistics.
Introductory topics covered deal with basic
concepts such as impulse response function, Wold decomposition, etc.
There is a good emphasis on graphical tools and characteristics of
generating good plots which are useful for practical analysis. The
section is rounded off by a discussion on bad graphics.
The
rest of the book is very well structured - starting with stationary
models and discussing details of the ARMA and checking for
stationarity. The part on non-stationary models deals clearly with
standard issues of detection, ARIMA.
Section on time series
model selection covers issues including the use of the AIC, bias
corrected information criteria and Bayesian information criteria. The
section also discusses comparing impulse response models for
competing models.
Students should find the example analysis
cases for a chemical process and seasonal company sales model very
helpful in the practical aspects of building time series models. The
practical computing example for impulse response function with a
spreadsheet helps to build confidence in model building.
Sufficient
depth is provided into issues with ARIMA model including constant
terms, over differencing and under differencing, and handling missing
values in time series models, covering most issues which are of
concern to basic to intermediate level analysis.
Transfer
function model discussion includes identifying the transfer function,
modeling the noise, general methodology and forecasting. Additional
topics such as spurious relationships, process regime changes, and
multiple time series are included towards the end of the book making
it a comprehensive guide.
The coverage of all the basic topics
with real world examples should appeal to a wide range of fields
where time series methods are used. While the book adopts a clear and
simplified approach the depth of issues is not compromised while
keeping an eye on implementation.
Time Series Analysis and
Forecasting by Example is well recommended as a great introductory
book for students transitioning from general statistics to time
series as well as a good source book for intermediate level time
series model builders.