Seasonal Adjustment – Quarterly Services Survey (QSS)
Frequently Asked Questions (FAQs)
Question: What is seasonal adjustment?
Answer: Seasonal adjustment is the process of estimating and removing seasonal effects from a time series in order to better reveal certain non-seasonal features. An example of a seasonal effect is an increase in accounting services during the tax season. (Seasonal effects are defined more precisely below.)
Question: Why do you seasonally adjust data?
Answer: Seasonal movements are often large enough that they mask other characteristics of the data that are of interest to analysts who track current economic trends. For example, if each quarter has a different seasonal tendency toward high or low values it can be difficult to detect the general direction of a time series' recent quarterly movement (increase, decrease, turning point, no change, consistency with another economic indicator, etc.). Seasonal adjustment produces data in which the values of neighboring quarters are usually easier to compare. Many data users prefer seasonally adjusted data because they want to see those characteristics that seasonal movements tend to mask, especially changes in the direction of the series.
Question: In the original (unadjusted) series, this year's 2nd quarter value is larger than the 1st quarter value. However, the seasonally adjusted series shows a decrease from 1st quarter to 2nd quarter this year. What does this discrepancy mean?
Answer: This difference in direction can happen only when the seasonal factor for 2nd quarter is larger than the seasonal factor for 1st quarter, indicating that when the underlying level of the series isn't changing, the 2nd quarter value will typically be larger than the 1st quarter value. This year, the original series' 2nd quarter increase over the 1st quarter value must be smaller than usual, either because the underlying level of the series is decreasing or because some special event or events abnormally increased the 1st quarter value somewhat, or decreased the 2nd quarter value somewhat.
Question: What kinds of seasonal effects are removed during seasonal adjustment?
Answer: Seasonal adjustment procedures for quarterly time series estimate effects that occur in the same calendar quarter with similar magnitude and direction from year to year. Examples of these effects include economic activity tied to the tax season or to the travel season. The seasonal factors are estimates of average effects for each quarter. It is important to note that seasonal factors are based on present and past experience and that future data may show a different pattern of seasonal factors.
Question: What is the seasonal adjustment process?
Answer: The mechanics of seasonal adjustment involve breaking down a series into trend-cycle, seasonal, and irregular components
- Trend-Cycle: Level estimate for each quarter derived from the surrounding year-or-two of observations.
- Seasonal Effects: Effects that are reasonably stable in terms of annual timing, direction, and magnitude.
Possible causes include natural factors (the weather), administrative measures (starting and ending dates of
the school year), and social/cultural/religious traditions (fixed holidays such as Christmas).
- Irregular Component: Anything not included in the trend-cycle or the seasonal effects. Its values are
unpredictable as regards timing, impact, and duration. It can arise from sampling error, non-sampling error,
unseasonable weather, natural disasters, strikes, etc.
Question: What is X-12-ARIMA?
Answer: X-12-ARIMA is a seasonal adjustment program developed at the U.S. Census Bureau. The program is based on the Bureau's earlier X-11 program and the X-11-ARIMA/88 program developed at Statistics Canada.
Improvements to X-12-ARIMA include:
- Use of ARIMA models to forecast the series, allowing us to use better, symmetric
moving averages that give us generally smaller revisions to the seasonal factors;
- New diagnostic tools; Wider variety of moving average options; and
- New user interface.
For more information see X-12-ARIMA
Question: How is the seasonal adjustment derived?
Answer: We use X-12-ARIMA to derive our seasonal adjustment and produce seasonal factors.
It is difficult to estimate seasonal effects when the underlying level of the series changes over time. For this reason, the program starts by detrending the series with a crude estimate of the trend-cycle. It then derives crude seasonal factors from the detrended series. It uses these to obtain a better trend-cycle and detrended series from which a more refined seasonal component is obtained. This iterative procedure, involving successive improvements, is used because seasonal effects make it difficult to determine the underlying level of the series required for the first step. Crude and more refined irregular components are used to identify and compensate for data that are so extreme that they can distort the estimates of trend-cycle and seasonal factors.
The seasonal factors are divided into the original series to get the seasonally adjusted series. For example, suppose for a particular quarter, a series has a value of 100,000 and a seasonal factor of 0.80. The seasonally adjusted value for this quarter is 100,000/0.80=125,000.
Question: Why do you not adjust for trading day and moving holiday effects?
Answer: The series are currently too short for X-12-ARIMA to accurately estimate trading day and moving holiday effects. Furthermore, trading day effects tend to be subtle in quarterly series.
Question: What is the difference between an additive model and a multiplicative model?
Answer: An additive model is used when the seasonal variations of a series are roughly constant in magnitude. The unadjusted data are not transformed prior to the seasonal adjustment process. A multiplicative model is preferred when the seasonal variations are roughly proportional to the level of the series. A log transformation is taken of the unadjusted data prior to the seasonal adjustment process.
Question: How do you determine which series to seasonally adjust?
Answer: Due to the shortness of the series, seasonally adjusted estimates will be published only for QSS series with indications of a good quality seasonal adjustment. However, research will continue on the seasonal adjustment of additional series.
Question: What indicates a good quality seasonal adjustment?
Answer: No residual seasonal effect. Once we adjust the series for seasonality, there should be no remaining seasonal effect in the adjusted series. The seasonally adjusted series is the combination of the trend-cycle and the irregular. Neither of these components should contain seasonality.
Passing values for quality assessment diagnostics. . We look for M7 less than 1.0 and statistically significant F-statistics. These diagnostics help us decide if X-12-ARIMA can adequately adjust the series.
Stability (small revisions) of the estimates. Due to the shortness of the series, limited X-12-ARIMA diagnostics are available to evaluate the stability of the series. X-12-ARIMA options are selected to best minimize revisions of the estimates. We can also reduce revisions by running X-12-ARIMA every quarter to get concurrent seasonal factors and by using ARIMA forecasts that make it possible to use symmetrical averaging formulas in the calculation of the seasonal factors, trend-cycle, and irregulars.
Question: What is concurrent seasonal adjustment?
Answer: Each quarter, concurrent seasonal adjustment uses all available unadjusted estimates (including the latest preliminary estimates) as input to the X-12-ARIMA program. Factors derived from the concurrent seasonal adjustment are applied to the unadjusted estimates for the most recent six quarters.
Question: Why do you revise seasonal factors?
Answer: There are two reasons that we revise seasonal factors:
- We revise factors when we revise the unadjusted data to achieve a better fit to the revised data.
- The estimate of a seasonal factor for a given quarter, say 1st quarter 2005, is most strongly influenced by the data from surrounding
1st quarters (especially from 2004 and 2006). In 2005, when the 1st quarter data for 2006 and later are not available, the seasonal
factor estimate for 1st quarter 2005 will be of reduced quality, unless X-12-ARIMA has calculated good forecasts of data for 2006
and later years and has used them in place of the data that is not yet available. In any case, when future data become available,
we use them to obtain improved seasonal factor estimates. These revised factors lead to revised seasonal adjustments of higher
quality.
We revise the seasonal factors for the previous five quarters because they are most affected when the preliminary quarter and revised previous quarter data are available. In the Annual Benchmark Report released 4th quarter each year, the seasonally adjusted estimates and seasonal factors will be revised for the entire series to reflect revisions in the unadjusted estimates based on the results of the Service Annual Survey.
Question: Why can't I get the annual total by summing the seasonally adjusted quarterly values?
Answer: When seasonal adjustment is done by dividing the time series by seasonal factors, it is arithmetically impossible for the adjusted series to have the same annual totals as the unadjusted series (except in the uninteresting case in which the time series values repeat perfectly from year to year). "Benchmarking" procedures can be used to modify the adjusted series so as to force the adjusted series to have the same totals as the unadjusted series, but these procedures do not account for evolving seasonal effects.
Question: What is an indirect adjustment? Why is it used?
Answer: If an aggregate time series is a sum (or other composite) of component series that are seasonally adjusted, then the sum of the adjusted component series provides a seasonal adjustment of the aggregate series that is called the indirect adjustment. This adjustment is usually different from the direct adjustment that is obtained by applying the seasonal adjustment program to the aggregate (or composite) series. When the component series have quite distinct seasonal patterns and have adjustments of good quality, indirect seasonal adjustment is usually of better quality. Indirect seasonal adjustments are preferred by many data users because they are consistent with the adjustments of the component series.
56pt is indirectly adjusted by summing seasonally adjusted estimates for 5613, 5615, and 562, and unadjusted estimates for 561pt*. Seasonal factors for 56pt are derived directly and can be used to approximate the published seasonally adjusted estimates at this level. All other published seasonally adjusted series for QSS are adjusted directly.
