Introductory Time Series Analysis Textbook

[hsu]

I'll be taking an introductory course on time series analysis in the spring, and we will be using the instructor's online notes as the "textbook". My previous experiences with such instructor's notes have been that they contain only the essentials of the course and aren't really useful as references. Besides, I always like using a physical textbook.

So, can anyone recommend a good textbook for this subject that is both clear for learning from and useful as a reference? The list of topics to be covered is:

Trend analysis, trend-based methods (moving averages, exponential smoothing), stationary processes, ARMA and ARIMA models, spectrum and its estimation, frequency filtration, seasonal models, multivariate models, Kalman filter.

By the way, another professor who sometimes teaches this course at my university has used this textbook:

https://www.amazon.com/dp/0321322169/?tag=pfamazon01-20

in the past; would that be one to consider?

Thanks!

 

[atyy]

I liked Shumway and Stoffer for time series, but for the Kalman filter Dan Simon's book is exceptionally friendly.

 

[tomdireill]

Hsu,

The Wei textbook is excellent for a number of reasons. It describes what to look for that other textbooks don't address. Chapter 9 discussed outliers and how important they are in order to model the data (ie additive outliers(AO), innovative outliers(IO)). Example 9.5 uses our software (Autobox) in the example. It doesn't discuss level shifts (ie 0,0,0,0,1,1,1,1,etc) or seasonal pulses (ie 0,1,0,0,0,0,0,0,0,0,0,0,0,1,0,0,etc for monthly data).

The transfer function models examples also used Autobox, but do not show the use of them with interventions as the book was written back in 1990 and did not include those advances over time (but what other textbook does? :) You can use Autobox as a student with ~700 time series from textbooks and of course Wei's.

See the ackowledgements section in the front for our name.

We would be happy to answer any questions here or off-line.

www.autobox.com

Regards
Tom Reilly