Cover Image
QR Code

Model-based seasonally adjusted estimates and sampling error.

The article presents a study which applied an experimental model-based method to selected Current Population Survey (CPS) demographic series. The presence of large survey errors in the detailed CPS series represents a major challenge to conventional methods of seasonal adjustment. The U.S. Bureau of...

Full description

Published in:Monthly Labor Review Vol. 128; no. 9; pp. 27 - 38
Main Authors: Tiller, Richard, Di Natale, Marisa
Format: Article
Published: US Department of Labor Sept2005
Subjects:
United States
United States. Bureau of Labor Statistics
Demographic surveys
Population
Mathematical models
Statistics
Online Access:View this record in EBSCOhost
Summary:The article presents a study which applied an experimental model-based method to selected Current Population Survey (CPS) demographic series. The presence of large survey errors in the detailed CPS series represents a major challenge to conventional methods of seasonal adjustment. The U.S. Bureau of Labor Statistics (BLS) uses a seasonal adjustment program called X-12-ARIMA to seasonally adjust its CPS series. An alternative that is gaining increasing attention is the model-based approach to seasonal adjustment. A comparison of the two approaches suggests that the model-based approach provides much-needed flexibility in controlling for the effects of sampling error. Such flexibility is not possible with the conventional moving-average approach.