New seasonal adjustment factors for household data series – Statistical Data Included
Richard B. Tiller
Semiannually, the Bureau of Labor Statistics publishes the factors to be used during the following 6 months for seasonal adjustment of the major labor force series. Table 1 presents the seasonal adjustment factors for the 12 major labor force components for the period July-December 2002. The factors for these components for the first 6 months of 2002 were published in the January 2002 issue of Employment and Earnings. The 12 seasonally adjusted series that remit from the application of these factors are used in the computation of the seasonally adjusted figures for the national overall levels of labor force, employment, and unemployment and for the overall unemployment rate.
The new seasonal factors have been extrapolated using the X-11 ARIMA program with data from January 1990 through June 2002 for each series. The ARIMA models used were the same as those used earlier in the year for the extrapolation of the factors for January through June; they were identified in the January 2002 issue of this publication. The historical seasonally adjusted data, including those for the first 6 months of 2002, will not be revised until the release of January 2003 data in February 2003.
Because of the changes introduced at the beginning of 1994, 1997, and 1999 in the survey and the processing procedures on which the labor force series estimates are based, (1) prior adjustment factors were used in these X-11 ARIMA runs to link the pre-1994, pre-1997, and/or pre-1999 data with the subsequent data for purposes of seasonal adjustment.
Without prior adjustment, those changes could have caused distortion in the seasonal decomposition. The pre 1994, pre-1997, and pre-1999 prior adjustment factors used in these runs were the same as those used in the runs that provided the factors for January through June. The prior adjustment factors for the 12 major components are shown in table 1 alongside the seasonal factors.
More detailed information on the seasonal adjustment procedures and the revision of historical data is included in “Revision of Seasonally Adjusted Labor Force Series,” Employment and Earnings, January 2002.
Data for any of the several hundred seasonally adjusted labor force series and the July-December 2002 seasonal adjustment factors for any of the other independently adjusted series–there are about 170 in addition to the 12 major components–may be obtained from BLS upon request. Requests for data or inquiries concerning seasonal adjustment methods or the availability of machine-readable files of other labor force data should be addressed to the Division of Data Development and Publications, Office of Employment and Unemployment Statistics, Bureau of Labor Statistics, Washington, DC 20212-0001.
Changes in Seasonal Adjustment Procedures
Due to the introduction of population controls from the 2000 Decennial Census as well as other changes being made to the Current Population Survey (CPS), seasonally adjusted CPS labor force estimates, which are usually revised each January with the release of December data, will not be revised until February 7, 2003, with the release of January 2003 data.
At the same time, the BLS will begin using the X-12-ARIMA program to seasonally adjust the CPS labor force estimates, replacing the X-11 ARIMA program which has been used since January 1980. The X-12-ARIMA program includes all of the capabilities of the X-11 ARIMA program while also introducing new features, Developed at the U.S. Census Bureau, X-12-ARIMA provides enhancements to (1) ARIMA time series modeling and model selection, (2) detection and estimation of outlier, trading day, and holiday effects, (3) post-adjustment diagnostics, and (4) seasonal and trend filter options. For a detailed description of the X-12-ARIMA program and its features, see D.F. Findley, B.C. Monsell, W.R. Bell, M.C. Otto, and B.C. Chen, “New Capabilities and Methods of the X-12-ARIMA Seasonal Adjustment Program,” Journal of Business and Economic Statistics, April 1998, Vol. 16, No. 2.