A short course on seasonal time series
Instructors: William Bell, David Findley, Tucker McElroy, and Brian Monsell of the U.S. Census Bureau
Dates: September 7–10
Location: Bureau of Labor Statistics, 2 Massachusetts Ave., NE, Washington, DC 20212
Registration: Register online
Payment: Online by credit card or mail by check, $275 or less
With the release of X-13ARIMA-SEATS scheduled for the end of 2010, it is important for practitioners of seasonal adjustment to have a basic understanding of the process for both model-based and X-11–based seasonal adjustment, as well as regARIMA modeling.
X-13ARIMA-SEATS is a collaboration between the U.S. Census Bureau and Bank of Spain. It combines the X-12-ARIMA package, developed at the bureau, with an implementation of the SEATS package for model-based seasonal adjustment, developed under the guidance of Agustin Maravall at the Bank of Spain.
The course will provide users with a sound methodological background for using the X-13ARIMA-SEATS program. It will alternate between theory and practice and include time for participants to use and explore the software with guidance from the instructors. Some of the topics to be covered include the following:
- ARIMA modeling and forecasting
- regARIMA modeling: adding regression terms to detect outliers and model calendar effects (such as trading-day and Easter) and intervention effects
- Underlying X-11 and SEATS processes
- Use and interpretation of seasonal adjustment and modeling diagnostics
- Using the U.S. Census Bureau’s Win X-12 interface to X-13-ARIMA-SEATS
Coursework in mathematical statistics is recommended, but anyone who currently uses X-12-ARIMA will benefit from taking this class, which is aimed at those who perform seasonal adjustment in their organizations.
Participants will receive a copy of all the slides, a set of notes on regARIMA modeling, and handouts on how to obtain the software and utilities used in the class.
The currently distributed versions of X-12-ARIMA, Win X-12, and X-12-Graph, as well as resources for new users, are available here.