Seasonal Adjustment in the CPI
Each year, with the release of the January CPI, seasonal adjustment
factors are recalculated to reflect price movements from the
just-completed calendar year. This routine recalculation may result in
revisions to seasonally adjusted indexes for the previous five years.
Basic information on the use of seasonal adjustment may be found in our
Fact Sheet on Seasonal
For a more technical discussion of seasonal adjustment methodology, see
the following excerpts from the BLS Handbook of Methods:
2015 Seasonal Factors and Seasonally Adjusted Data Revisions for January 2015, issued February 2015
A Note on the Use of Seasonally Adjusted and Unadjusted Data
Because price data are used for different purposes by different groups,
the Bureau of Labor Statistics publishes seasonally adjusted as well as
unadjusted changes each month.
For analyzing general price trends in the economy, seasonally adjusted
changes are usually preferred since they eliminate the effect of changes
that normally occur at the same time and in about the same magnitude every
year—such as price movements resulting from changing climatic conditions,
production cycles, model changeovers, holidays, and sales.
The unadjusted data are of primary interest to consumers concerned
about the prices they actually pay. Unadjusted data also are used
extensively for escalation purposes. Many collective bargaining contract
agreements and pension plans, for example, tie compensation changes to the
Consumer Price Index before adjustment for seasonal variation.
Seasonal factors used in computing the seasonally adjusted indexes are
derived by the X-13ARIMA-SEATS Seasonal Adjustment Method. Seasonally adjusted
indexes and seasonal factors are computed annually. Each year, the last five
years of seasonally adjusted data are revised. Data from January 2010
through December 2014 were replaced in January 2015. Exceptions to the
usual revision schedule were: the updated seasonal data at the end of 1977
replaced data from 1967 through 1977; and, in January 2002, dependently
seasonally adjusted series were revised for January 1987-December 2001 as
a result of a change in the aggregation weights for dependently adjusted
series. For further information, please see "Aggregation of Dependently
Adjusted Seasonally Adjusted Series" (PDF).
Effective with the publication of data from January 2006 through
December 2010 in January 2011, the Video and audio series and the
Information technology, hardware and services series were changed from
independently adjusted to dependently adjusted. This resulted in an
increase in the number of seasonal components used in deriving seasonal
movement of the All items and 64 other lower level aggregations, from 73
for the publication of January 1998 through December 2005 data to 82 for
the publication of seasonally adjusted data for January 2006 and later.
Each year the seasonal status of every series is reevaluated based upon
certain statistical criteria. If any of the 82 components change their
seasonal adjustment status from seasonally adjusted to not seasonally
adjusted, not seasonally adjusted data will be used in the aggregation of
the dependent series for the last five years, but the seasonally adjusted
indexes before that period will not be changed. Note: 32 of the 82
components are not seasonally adjusted for 2015.
Seasonally adjusted data, including the All items index levels, are
subject to revision for up to five years after their original release. For
this reason, BLS advises against the use of these data in escalation
Effective with the calculation of the seasonal factors for 1990, the
Bureau of Labor Statistics has used an enhanced seasonal adjustment
procedure called Intervention Analysis Seasonal Adjustment for some CPI
series. Intervention Analysis Seasonal Adjustment allows for better
estimates of seasonally adjusted data. Extreme values and/or sharp
movements that might distort the seasonal pattern are estimated and
removed from the data prior to calculation of seasonal factors. Beginning
with the calculation of seasonal factors for 1996, X-12-ARIMA software was
used for Intervention Analysis Seasonal Adjustment. For more information,
see "Improvements to
CPI Procedures for Intervention Analysis Seasonal Adjustment" (PDF),
which was originally published in the December 1996 CPI Detailed
Report. In 2015, for the 2010-2014 revisions, the Bureau of Labor Statistics began using X-13ARIMA-SEATS to
perform seasonal adjustment of CPI series, including Intervention Analysis seasonal adjustment for certain series.
For the seasonal factors introduced in January 2015, BLS adjusted 33
series using Intervention Analysis Seasonal Adjustment, including selected
food and beverage items, motor fuels, electricity and vehicles. For
example, this procedure was used for the Motor fuel series to offset the
effects of events such as the response in crude oil markets to the worldwide economic downturn in 2008.
For a complete list of series that used Intervention Analysis Seasonal
Adjustment, see "Intervention Analysis
Seasonal Adjustment" (PDF).
For additional information on seasonal adjustment in the CPI, please
write to the Bureau of Labor Statistics, Division of Consumer Prices and
Price Indexes, Washington, DC 20212 or contact Christopher Graci or Carlyle Jackson on (202)
691-6968 or by e-mail at Graci.Christopher@bls.gov or Jackson.Carlyle@bls.gov. If you have
general questions about the CPI, please call our information staff at
Last Modified Date: February 20, 2015