A substantial revision of the Census Bureau’s X-11 seasonal adjustment program is underway. The partially completed version of the new program presently being evaluated contains important new diagnostics and the capability of fitting ARIMA models or regressions with ARIMA errors by “nearly” exact maximum likelihood estimation. The “sliding spans” diagnostics represent a particularly valuable addition to the program’s diagnostics. This paper illustrates the application of these to questions of seasonal and trading day adjustability, seasonal filter length choice, direct versus indirect adjustment, and seasonal adjustments versus trends. Some of the inadequacies in X-11’s F-statistics revealed by these analyses are confirmed by using the minimum AIC procedure to compare regression-with-ARIMA-errors models. Such comparisons, which the program facilitates, also reveal the importance of outlier models. The use of outlier models reveals that the S-11 procedure is less outlier resistant than might have been expected.

Others in Series

Working Paper

Construction of a Housing Index

November 21, 1988

Construction of a Housing Index

Working Paper

Variance Formulae for the Generalized Composite Estimator Under a B...

December 27, 1988

Variance Formulae for the Generalized Composite Estimator Under a Balanced One-level Rotation Error

Working Paper

Modeling Time Series Subject to Sampling Error

January 03, 1989

Modeling Time Series Subject to Sampling Error

View All

Related Information

WORKING PAPER

Statistical Research Reports and Studies

Page Last Revised - October 28, 2021

Some content on this site is available in several different electronic formats. Some of the files may require a plug-in or additional software to view.

Is this page helpful?
Thumbs Up Image

Yes

NO THANKS

255 characters maximum

255 characters maximum reached

Thank you for your feedback.
Comments or suggestions?

Top