Seasonal adjustment is used to identify and factor out seasonal movements in some Consumer Price Index (CPI) series. The resultant seasonally adjusted data reflect an estimate of non-seasonal price movements. The CPI uses the Census Bureau’s X-13ARIMA-SEATS software to calculate factors for seasonal adjustment of both historical and current series.
Seasonal adjustment for certain CPI series includes a technique called intervention analysis seasonal adjustment (IASA) that is part of the Census X-13 seasonal adjustment program. The seasonal factors produced for CPI series using IASA more accurately represent the underlying seasonal pattern, because it allows economic phenomena that are not seasonal in nature, such as outliers and level shifts, to be factored out of indexes before the calculation of seasonal factors. An outlier is an extreme value for a particular month. A level shift is a change or shift in the price level of a CPI series caused by an event, such as a sales tax increase or oil embargo, occurring over one or several months. The result is an adjustment based on a representation of the series with the seasonal pattern emphasized. IASA also makes it possible to account for seasonal shifts, resulting in a better seasonal adjustment in the periods before and after the shift occurred.
X-13ARIMA-SEATS’s built-in regression variables are used for directly estimating the effects of sudden level changes and other disruptions and removing those effects before the calculation of seasonal factors. Seasonal factors are applied to the original unadjusted series without interventions removed. As a result, level shifts and outliers, removed for the calculation of seasonal factors, are present in the seasonally adjusted series.
A CPI series may receive IASA if the series is directly adjusted and has a relative importance greater than 0.5% to the U.S. city average All items index. If IASA is used for a component of the seasonally adjusted U.S. city average All items index, series which are subsets of the component series are also eligible for IASA. For a comprehensive discussion of IASA, see "Improvements to CPI Procedures for Intervention Analysis Seasonal Adjustment" in the December 1996 issue of the CPI Detailed Report.
This year, BLS examined CPI series using the 8-year span from January 2014 through December 2021 with some exceptions. The following series were examined using a 7-year span from January 2015 through December 2021: used cars and trucks, new vehicles, utilities and public transportation and women’s footwear. In 2018, the Used cars and trucks index changed its estimation methodology from using a 3-month moving average of depreciated prices to a 1-month price change approach. Seasonal factors for this series were determined using six years of estimated 1-month price change.
In January 2022, BLS adjusted the series listed below using IASA techniques:
For each eligible series that was adjusted using X-13ARIMA-SEATS intervention analysis seasonal adjustment, a list of level shifts is provided in the file below, along with the identified causes (events). Outliers are also included in the file.
For additional information on seasonal adjustment in the CPI, please contact us at (202) 691-6968 or send an e-mail to firstname.lastname@example.org.
Last Modified Date: February 8, 2022