Statistics 451 - Applied Time Series
Spring 2016

Instructor: William Q. Meeker, 294-5336
Instructor email:
Office Hours: WF 2:10-3:00 p.m. in 2109 Snedecor Hall (or by appointment---send email)
Lecture: MWF 12:10-1:00, Gilman 2205
Assignments: Due weekly on Fridays, unless announced otherwise.
Examinations: Two one-hour exams plus a two-hour final examination.

Statistics 451 homepage:
Text: None required; see suggested reading below
Class Notes: Available from the course webpage.

Tentative Schedule

  1. Introduction to time series and time series models. (Module 1 and Chapters 1 and 2, 2 days)
  2. Review of multiple regression and descriptive methods, and further introduction to time series analysis. (Module 2; 6 days) Graphical techniques, transformations, analysis of residuals, autocorrelation.
  3. Fundamental concepts of time series modleing. (Module 3, Chapters 1 and 3; 10 days) Stochastic processes, process and sample autocorrelation function, partial autocorrelation function.
  4. Stationary time series models. (Modules 4 and 5, Chapter 3; 6 days) ARIMA models, model identification. Nonstationary time series. (Module 6, Chapter 4; 3 days) Differencing, transformation, identification.
  5. Estimation in time series. (Module 7; 2 days) Nonlinear least squares, potential problems, diagnostic checking and residual analysis.
  6. Forecasting and prediction intervals. (Module 8, Chapters 4 and 5; 4 days)
  7. Seasonal time series models. (Module 9, Chapter 5; 3 days) SARIMA models, identification, estimation, forecasting.
  8. Transfer function and intervention models. (Modules 10 and 11, Chapters 8; 11 days)
  9. Vector (multivariate) time series. (Module 12, notes; 3 days)

Learning outcomes:
Upon successful completion of this course, students will understand the nature of time series data and the structure of various time series models for stationaty, nonstationary, and seaconal time series. They will be able to identify an appropriate time series model, fit such models to data, and use the models for purposes of forecasting, including quantification of forecast errors through the construction of prediction intervals. Students will understand how to extend univariate models to allow for the use of explanatory varialbes to model both interventions and transfer functions.


Recommended Reading:
Time Series Analysis and Forecasting by Example by Soren Bisgaard and Murat Kulahci, John Wiley and Sons.
Time series analysis and its applications. Shumway, R. H., & Stoffer, D. S. (2013). Springer Science & Business Media.
Time Series Analysis - Univariate and Multivariate Methods (Second Edition) by William W.S.Wei

Supplementary Reading:
Time Series Analysis: Forecasting and Control by G.E.P. Box and G.M. Jenkins
Practical Experiences with Modeling and Forecasting Time Series by G.M. Jenkins

Students with disabilities
Please address any special needs or special accommodations with me at the beginning of the semester or as soon as you become aware of your needs. Those seeking accommodations based on disabilities should obtain a Student Academic Accommodation Request (SAAR) form from the Disability Resources (DR) office (515-294-6624). DR is located on the main floor of the Student Services Building, Room 1076.

Click here to go to W.Q. Meeker's homepage.