At the end of each calendar year, BLS updates the seasonal adjustment factors to incorporate the data for that year and revises the historical series for the prior four years. This year, seasonally adjusted data for January 2008-November 2012 were revised. The revised data replace the previously published seasonally adjusted data and are available on the Web site's time series database. The tables below summarize the changes resulting from these revisions.
BLS seasonally adjusts six monthly MLS series using the X-12-ARIMA program. Seasonal adjustment is the process of estimating and removing periodic fluctuations caused by events such as weather, harvests, holidays, and the opening and closing of schools. The purpose for seasonal adjustment is to make it easier to observe fundamental changes in the level of the series, particularly those associated with general economic expansions and contractions. The method of concurrent seasonal adjustment is performed monthly in order to use all available data, including the current month.
Prior adjustments are applied to the series to adjust the original data for differences in the number of weeks used to calculate monthly data. As weekly unemployment insurance claims filings are aggregated to form monthly data, a particular months value could be calculated with five weeks of data in one year and four weeks in another. This can seriously distort the seasonal factors if ignored in the seasonal adjustment process. These effects are modeled in the X-12-ARIMA program and permanently removed from the final seasonally adjusted series.
In addition, large, unexpected, and rare events such as Hurricane Katrina can cause a time series to deviate from its expected patterns. When these types of events occur, changes are made to the MLS seasonal adjustment procedure through the use of intervention variables to keep the seasonal factors from becoming seriously distorted.
Last Modified Date: January 25, 2013