An official website of the United States government

The .gov means it's official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Bureau of Labor Statistics > Current Employment Statistics (State and Metro Area) > Methods > Seasonal Adjustment

Research Data on Concurrent Seasonal Adjustment by State and Benchmark Year

Overview
The Current Employment Statistics (CES) State and Area program developed research data on concurrent seasonal adjustment for states and Metropolitan Statistical Areas (MSAs). The CES program simulated concurrent seasonally adjusted data for three years and is now providing this data to the public. .

The data is available for benchmark years 2014, 2015 and 2016. For each benchmark year, the research concurrent data is provided along with the annual forecasted seasonally adjusted data previously published by CES. The data is available by state for each published seasonally adjusted series within a specified benchmark year.

Concurrent vs Forecast Methodology
In 2018, CES converted to concurrent seasonal adjustment for all states and MSAs. This new methodology uses all available estimates, including those for the current month, in the development of seasonal factors. Previously, the program forecasted seasonal factors once a year during the annual seasonal adjustment process.

The annual seasonal adjustment process requires 10 years of historical sample data as an input to the X-13 ARIMA model to create forecasted factors. These factors are then used to seasonally adjust sample estimates for the remainder of the year. The concurrent seasonal adjustment process uses the same historical sample data. Therefore, the ARIMA model, outliers, and calendar effects determined during the annual review process are used in concurrent seasonal adjustment. The only difference in inputs between the two methodologies is the incorporation of real-time estimates with concurrent seasonal adjustment.

For additional information regarding the transition to concurrent seasonal adjustment, please see https://www.bls.gov/sae/seasonal-adjustment/research-data-on-concurrent-seasonal-adjustment-by-state-and-benchmark-year.htm

Notes on the Datasets
The data is available in Excel file format, one for each benchmark year for a total of three. Once a benchmark year is selected, the data user can review either the full set of concurrent seasonal adjustment data or a particular state individually. The user can review the full set of data by selecting the “ Total” tab on the Excel file or a particular state by selecting that state’ s corresponding tab.

The first monthly observation for all series will be the September of the previous calendar year. The seasonally adjusted data for this observation is calibrated using a seasonal factor derived during annual production and will, therefore, be the same for both the concurrent and forecasted time series. This month should serve as a constant starting point when comparing the two datasets.

The datasets each have nine columns and the variables are defined below:

Benchmark Year: The benchmark year for which the data is estimated. The benchmark year classifies the timeframe in which the historical CES All Employees (AE) data, derived from estimation methods, is replaced with universe data derived mainly from the Quarterly Census of Employment and Wages (QCEW) program. The seasonally adjusted data published on these tables are the monthly estimated data derived following the replacement period in the benchmark year.

For example, benchmark year 2014 spans from April 2013 to December 2015. The employment data from April 2013 to September 2014 have been replaced with universe data while October 2014 to December 2015 is considered the current estimation timeframe. The research concurrent data is available for every month within the current estimation timeframe.

Series ID: This 20 digit identification code corresponds to the CESSA series code that classifies all published series. For more information on the Series ID code, please refer to https://www.bls.gov/help/hlpforma.htm#SM

State: The state for which the data is estimated. The research concurrent data is available for all 50 states plus the District of Columbia, Puerto Rico and the U.S. Virgin Islands.

Area: The area for which the data is estimated. The research concurrent data is available for all Metropolitan Statistical Areas (MSAs) that published seasonally adjusted data in the corresponding benchmark year.

Industry: The 2017 NAICS industry or employment sector for which the data is estimated. The research concurrent data is available for all industries that published seasonally adjusted data in the corresponding benchmark year. Most statewide series publish seasonally adjusted data at the extended super sector level, while MSA data is only published seasonally adjusted at the Total Nonfarm level.

Year: The reference year for which the data is estimated.

Month: The reference month for which the data is estimated.

Concurrent Seasonal Adjusted Data: The simulated concurrent seasonally adjusted data. The data is rounded to hundreds and published in thousands.

Published Seasonal Adjusted Data: The annual forecasted seasonally adjusted data for the corresponding benchmark year. This seasonally adjusted data is what was originally published during that time.

Data Files

Benchmark Year 2014

Historical Concurrent Data 2014

Benchmark Year 2015

Historical Concurrent Data 2015

Benchmark Year 2016

Historical Concurrent Data 2016

Last Modified Date: April 4, 2018